No abstract

No abstract

No abstract

Zirconium and Hf are nearly identical geochemically, and therefore most of the crust maintains near-chondritic Zr/Hf ratios of ∼35–40. By contrast, many high-silica rhyolites and granites have anomalously low Zr/Hf (15–30). As zircon is the primary reservoir for both Zr and Hf and preferentially incorporates Zr, crystallization of zircon controls Zr/Hf, imprinting low Zr/Hf on coexisting melt. Thus, low Zr/Hf is a unique fingerprint of effective magmatic fractionation in the crust. Age and compositional zonation in zircons themselves provide a record of the thermal and compositional histories of magmatic systems. High Hf (low Zr/Hf) in zircon zones demonstrates growth from fractionated melt, and Ti provides an estimate of temperature of crystallization (TTiZ) (Watson and Harrison, 2005). Whole-rock Zr/Hf and zircon zonation in the Spirit Mountain batholith, Nevada, document repeated fractionation and thermal fluctuations. Ratios of Zr/Hf are ∼307–40 for cumulates and 18–30 for high-SiO2 granites. In zircons, Hf (and U) are inversely correlated with Ti, and concentrations indicate large fluctuations in melt composition and TTiZ (>100°C) for individual zircons. Such variations are consistent with field relations and ion-probe zircon geochronology that indicate a >1 million year history of repeated replenishment, fractionation, and extraction of melt from crystal mush to form the low Zr/Hf high-SiO2 zone.

A B S T R A C TThe Avila batholith of central Spain is composed of upper Carboniferous peraluminous granitoids that were preceded by volumetrically insignificant bodies of mafic-ultramafic hybrid magmas and postdated by several dike swarms of camptonitic lamprophyres. Rb-Sr dating indicates continuous magmatic activity from 350 Ma to 280 Ma, starting ∼ ∼ with the mafic precursors and a few midcrustal anatectic leucogranites, followed by massive autochthonous and allochthonous granodiorites and granites, and ending with the camptonitic lamprophyres. Early hybrid mafic magmas ( ; ) were produced in small batches during or immediately after the main deformation 340 Ma 340 Ma Sr ∼ 25 Nd ∼ Ϫ1.5 phase, probably by the partial melting of a mixture of 8 : 2 mantle and biotite-bearing crustal rocks at the crust-∼ mantle interface. These magmas were emplaced in the middle crust at 340 Ma, advecting a negligible amount of ∼ heat. The generation of crustal granites during the main deformation phase was very scarce and limited to highly fertile protoliths, rich in heat-producing elements, affected by strong shear zones. The generation of crustal granitoids on a batholithic scale took place from 330 Ma to 290 Ma, during the main extensional period. Granites ∼ ∼ ( -150; ) were produced by the partial melting of fertile crustal rocks 310 Ma 310 Ma Sr ∼ 45 Nd ∼ Ϫ2.1 to Ϫ 9 ( -218; ), characterized by high heat production ( 2.5-3 mW m Ϫ3 ). The zone of 310 Ma 310 Ma Sr ∼ 48Nd ∼ Ϫ2.2 to Ϫ 9 ∼ partial melting, 15-22 km in depth, was heated by thermal conduction from below after crustal thinning, but the ∼ contribution of radiogenic heat and the fertility of source rocks would have been essential for anatexis. The fast thinning of the crust from 310 Ma to 285 Ma released lithostatic pressure in the upper mantle and caused decom-∼ ∼ pressional melting of the metasome layer at 60-85 km in depth, producing camptonitic melts dated at 283 Ma. ∼ ∼ The existence of a fertile metasome layer implies that the lithospheric mantle beneath central Iberia was not actively involved in subduction during the Variscan orogeny.

No abstract

Dating the pre-Middle Ordovician metavolcanic rocks and metagranites of the Ollo de Sapo Domain has, historically, been difficult because of the small compositional variation, the effects of the Variscan orogeny and, as revealed in this paper, the unusually high fraction of inherited zircon components. The first reliable zircon data (U-Pb ion microprobe and Pb-Pb stepwise evaporation) indicate that the Ollo de Sapo volcanism spanned 495 ± 5 Ma to 483 ± 3 Ma, and was followed by the intrusion of high-level granites from 483 ± 3 Ma to 474 ± 4 Ma. In both metavolcanic rocks and metagranites, no less than 70-80 % of zircon grains are either totally Precambrian or contain a Precambrian core overgrown by a Cambro-Ordovician rim. About 80-90 % of inherited zircons are Early Ediacaran (602-614 Ma) and derived from calc-alkaline intermediate to felsic igneous rocks generated at the end of the Pan-African arc-continent collision. In the Villadepera region, located to the west, both the metagranites and metavolcanic rocks also contain Meso-Archaean zircons (3.0-3.2 Ga) which ultimately originated from the West African Craton. In the Hiendelaencina region, located to the east, both the metagranites and metavolcanic rocks lack Meso-Archaean zircons, but they have two different inherited zircon populations, one Cryogenian (650-700 Ma) and the other Tonian (850-900 Ma), which suggest older-than-Ediacaran additional island-arc components. The different proportion of source components and the marked variation of the 87 Sr/ 86 Sr init. suggest, at least tentatively, that the across-arc polarity of the remnants of the Pan-African arc of Iberia trended east-west (with respect to the current coordinates) during Cambro-Ordovician times, and that the passive margin was situated to the west.

The results of ~4000 LA-ICP-MS analyses in 152 thin sections from common crustal rocks reveal that several rock-forming minerals contain tens to a few thousand ppm of Zr. The highest concentrations of Zr are found in xenotime, followed by titanite, ilmenite, rutile, allanite, amphibole, clinopyroxene, garnet, magnetite and, less commonly, plagioclase, K-feldspar and orthopyroxene. Olivine, cordierite, biotite, muscovite, apatite, epidote and monazite have low levels of Zr (<5 ppm, generally <1 ppm). The minerals with the highest K D Hf /K D Zr are titanite (2.5), orthopyroxene (2.0), amphibole and clinopyroxene (1.8), and epidote and rutile (1.6-1.7). Ilmenite, magnetite, the feldspars and apatite have K D Hf /K D Zr ≈ 1, and values less than one were found in xenotime and zircon (0.8), garnet (0.7), and allanite (0.6). The most important implications of these results follow. First, the growth of a Zr-bearing phase during partial melting does not influence the Zr concentration of the melt, but increases the fraction of zircon that can be dissolved at a given temperature. This accelerates the disappearance of zircon from the protolith or, in melts already segregated, the dissolution of inherited zircon. The effect will be more marked in metaluminous magmas precipitating amphibole and titanite than in any other type of magma. Second, the presence of Zr-bearing phases has little influence of the zircon-saturation "geothermometer". It may cause somewhat higher (20-30°C) results in metaluminous rocks, but has no effect on peraluminous rocks. Third, the uptake of Zr by major minerals in crystallizing magmas may reduce both the concentration of Zr in the melt available to form zircon and the temperature at which zircon begins to precipitate. Mineral-melt reactions involving Zr-bearing phases may lead to zircon grains with complicated patterns of zoning and texturally discordant zones, apparently diachronous. Fourth, higher-than-chondrite Zr/Hf fractionates may arise from titanite, amphibole or clinopyroxene fractionation, but this requires very little or no crystallization of zircon. Significantly lower-than-chondrite Zr/Hf magmas only result from zircon fractionation. Lastly, two new examples of mineral reactions that involve the formation of a mass-balancing accessory phase, useful for high-resolution geochronology, are described: the formation of xenotime as a product of the breakdown of garnet in amphibolite-grade metapelites, and the subsolidus growth of a new rim on zircon included in Zr-bearing feldspars.