Gem corundum deposits of Madagascar: A review

Madagascar is one of the most important gem-producing countries in the world, including ruby and sapphires. Gem corundum deposits formed at different stages in the geological evolution of the island and in contrasting environments. Four main settings are identified: (1) Gem corundum formed in the Precambrian basement within the Neoproterozoic terranes of southern Madagascar, and in the volcano-sedimentary series of Beforona, north of Antananarivo. In the south, high-temperature (700 to 800 °C) and low-pressure (4 to 5 kbar) granulites contain deposits formed during the Pan-African orogenesis between 565 and 490 Ma. They accompany mafic and ultramafic complexes (ruby deposits of the Vohibory group), skarns at the contact between Anosyan granites and the Proterozoic Tranomaro group (sapphire deposits of the Tranomaro–Andranondambo district), and shear-zone corridors cross-cutting feldspathic gneisses, cordieritites and clinopyroxenites in the Tranomaro, Vohimena and Androyan metamorphic series (biotite schist deposits of Sahambano and Zazafotsy, cordieritites of Iankaroka and Ambatomena). The circulation of fluids, especially along discontinuities, allowed in-situ alkaline metasomatism, forming corundum host rocks related to desilicified granites, biotitites, “sakenites” and “corundumites”. (2) Gem corundum also occurs in the Triassic detrital formations of the Isalo group, as giant palaeoplacers in the Ilakaka–Sakaraha area. Here, sapphires and rubies may come from the metamorphic granulitic terranes of southern Madagascar. (3) Gem corundum deposits occur within the Neogene-Quaternary alkali basalts from Ankaratra (Antsirabe–Antanifotsy area) and in the Ambohitra Province (Nosy Be, Ambato and Ambondromifehy districts). Primary deposits are rare, except at Soamiakatra where ruby in gabbroic and clinopyroxenite xenoliths within alkali-basalts probably derive from mantle garnet peridotites. The blue-green-yellow sapphires typical of basaltic fields are always recovered in palaeoplacer (in karst formed upon Jurassic limestones from the Montagne d'Ambre, Antsiranana Province) and alluvial and soil placers (Ankaratra volcanic massif). (4) Deposits occur within Quaternary eluvial, colluvial and alluvial concentrations, such as high-quality rubies from the Andilamena and Vatomandry deposits.     

A preliminary stable isotope study on Mogok Ruby, Myanmar

The primary occurrence of ruby in the Mogok area, northern Myanmar is exclusively found in marble along with spinel–forsterite-bearing marble and phlogopite–graphite marble. These marble units are enclosed within banded biotite–garnet–sillimanite–oligoclase gneisses. Samples of these marbles collected for C–O stable isotope analysis show two trends of δ¹³C–δ¹⁸O variation resulting most likely from fluid–rock interactions. Ruby-bearing marble and phlogopite–graphite marble follow a trend with coupled C–O depletion, whereas spinel–forsterite-bearing marble follows a δ¹⁸O depletion trend with relatively constant δ¹³C values. Ruby formation might have resulted from CO₂-rich fluid–rock interaction, while spinel–forsterite-bearing marble was genetically related to CO₂-poor fluid–rock interaction. Both fluids may have arisen from external sources. Based on graphite Raman spectral thermometry, the estimated temperature for phlogopite–graphite marble, and probably ruby-bearing marble, was lower than 607 °C, and for spinel–forsterite-bearing marble, lower than 710 °C. Contrasting C/O diffusion between graphite/ruby/spinel/forsterite and calcite, local variations of isotopic compositions of newly formed minerals as a result of non-pervasive fluid infiltration, and open-system isotopic disturbance during cooling may have affected C-/O-isotopic fractionations between minerals. The estimated high formation temperatures for ruby and spinel/forsterite imply that the parental fluids may have been related to nearby igneous intrusions and/or metamorphic processes. Whether these two types of fluid were genetically related is unclear based on the present data.     

Tracing high-pressure metamorphism in marbles: Phase relations in high-grade aluminous calcite–dolomite marbles from the Greek Rhodope massif in the system CaO–MgO–Al2O3–SiO2–CO2 and indications of prior aragonite

Four different types of parageneses of the minerals calcite, dolomite, diopside, forsterite, spinel, amphibole (pargasite), (Ti–)clinohumite and phlogopite were observed in calcite–dolomite marbles collected in the Kimi-Complex of the Rhodope Metamorphic Province (RMP). The presence of former aragonite can be inferred from carbonate inclusions, which, in combination with an analysis of phase relations in the simplified system CaO–MgO–Al₂O₃–SiO₂–CO₂ (CMAS–CO₂) show that the mineral assemblages preserved in these marbles most likely equilibrated at the aragonite–calcite transition, slightly below the coesite stability field, at ca. 720 °C, 25 kbar and a_CO2 ~ 0.01. The thermodynamic model predicts that no matter what activity of CO₂, garnet has to be present in aluminous calcite–dolomite-marble at UHP conditions.     

P–T–t conditions of high-grade metamorphic rocks of the Garzon Massif, Andean basement, SE Colombia

The metamorphic evolution of the Garzón Massif, Colombia, is established on the basis of the textural, goethermobarometric, and geochronological relationships of the metamorphic minerals. The geothermobarometric data define a clockwise, nearly isothermal decompression path (ITD) for rocks from Las Margaritas migmatites, constrained by four P–T areas: 780–826 °C and 6.3–8.0 kbar, 760–820 °C and 8.0–8.8 kbar, 680–755 °C and 6.6–9.0 kbar, and 630 °C and 4 kbar. For the a garnet-bearing charnockitic gneiss from the Vergel granulites, the path is counterclockwise, constrained by geothermobarometric data of 5.3–6.2 kbar and 700–780 °C and 6.2–7.2 kbar and 685–740 °C. The clockwise ITD path represents a loop followed by the orogen during the transitional granulite–amphibolite metamorphic conditions, probably associated with a subduction process followed by a collisional tectonic event. This subduction framework produced continental crust thickening between 1148 and 1034 Ma and later collision with another continental block approximately 1000 Ma ago. The orogenic exhumation occurred with moderate uplift rate. The counterclockwise trajectory and two metamorphic events suggest a vertical displacement between the Vergel granulites and Las Margaritas migmatites units, because there is no isotopic difference that indicates the existence of different terranes. The data confirm that the metamorphic evolution for this domain was more dynamic than previously believed and includes: (1) metamorphic processes with the generation of new crust with a possible mixture of old material and (2) metamorphic recycling of continental crust. These geological processes characterize a complex Mesoproterozoic orogenic event that shares certain features with the Grenvillian basement rocks participating in the formation of Rodinia.     

Ultrahigh-temperature metamorphism in the Achankovil Zone: Implications for the correlation of crustal blocks in southern India

The Achankovil Zone of southern India, a NW–SE trending lineament of 8–10 km in width and > 100 km length, is a kinematically debated crustal feature, considered to mark the boundary between the Madurai Granulite Block in the north and the Trivandrum Granulite Block in the south. Both these crustal blocks show evidence for ultrahigh-temperature metamorphism during the Pan-African orogeny, although the exhumation styles are markedly different. The Achankovil Zone is characterized by discontinuous strands of cordierite-bearing gneiss with an assemblage of cordierite + garnet + quartz + plagioclase + spinel + ilmenite + magnetite ± orthopyroxene ± biotite ± K-feldspar ± sillimanite. The lithology preserves several peak and post-peak metamorphic assemblages including: (1) orthopyroxene + garnet, (2) perthite and/or anti-perthite, (3) cordierite ± orthopyroxene corona around garnet, and (4) cordierite + quartz symplectite after garnet. We estimate the peak metamorphic conditions of these rocks using orthopyroxene-bearing geothermobarometers and feldspar solvus which yield 8.5–9.5 kbar and 940–1040 °C, the highest P–T conditions so far recorded from the Achankovil Zone. The retrograde conditions were obtained from cordierite-bearing geothermobarometers at 3.5–4.5 kbar and 720 ± 60 °C. From orthopyroxene chemistry, we record a multistage exhumation history for these rocks, which is closely comparable with those reported in recent studies from the Madurai Granulite Block, but different from those documented from the Trivandrum Granulite Block. An evaluation of the petrologic and geochronologic data, together with the nature of exhumation paths leads us to propose that the Achankovil Zone is probably the southern flank of the Madurai Granulite Block, and not a unit of the Trivandrum Granulite Block as presently believed. Post-tectonic alkali granites that form an array of “suturing plutons” along the margin of the Madurai Granulite Block and within the Achankovil Zone, but are absent in the Trivandrum Granulite Block, suggest that the boundary between the Madurai Granulite Block and the Trivandrum Granulite Block might lie along the Tenmalai shear zone at the southern extremity of the Achankovil Zone.     

Geochronological problems related to polymetamorphism in the Limpopo Complex, South Africa

The integration of new and published geochronologic data with structural, magmatic/anatectic and pressure–temperature (P–T) process information allow the recognition of high-grade polymetamorphic granulites and associated high-grade shear zones in the Central Zone (CZ) of the Limpopo high-grade terrain in South Africa. Together, these two important features reflect a major high-grade D₃/M₃ event at ~ 2.02 Ga that overprinted the > 2.63 Ga high-grade Neoarchaean D₂/M₂ event, characterized by SW-plunging sheath folds. These major D₂/M₂ folds developed before ~ 2.63 Ga based on U–Pb zircon age data for precursors to leucocratic anatectic gneisses that cut the high-grade gneissic fabric. The D₃/M₃ shear event is accurately dated by U–Pb monazite (2017.1 ± 2.8 Ma) and PbSL garnet (2023 ± 11 Ma) age data obtained from syntectonic anatectic material, and from sheared metapelitic gneisses that were completely reworked during the high-grade shear event. The shear event was preceded by isobaric heating (P = ~ 6 kbar and T = ~ 670–780 °C), which resulted in the widespread formation of polymetamorphic granulites. Many efforts to date high-grade gneisses from the CZ using PbSL garnet dating resulted in a large spread of ages (~ 2.0–2.6 Ga) that reflect the polymetamorphic nature of these complexly deformed high-grade rocks.     

Metallogenesis of the Carajás Mineral Province, Southern Amazon Craton, Brazil: Varying styles of Archean through Paleoproterozoic to Neoproterozoic base- and precious-metal mineralisation

The Itacaiúnas Belt of the highly mineralised Carajás Mineral Province comprises ca. 2.75 Ga volcanic rocks overlain by sedimentary sequences of ca. 2.68 Ga age, that represent an intracratonic basin rather than a greenstone belt. Rocks are generally at low strain and low metamorphic grade, but are often highly deformed and at amphibolite facies grade adjacent to the Cinzento Strike Slip System. The Province has been long recognised for its giant enriched iron and manganese deposits, but over the past 20 years has been increasingly acknowledged as one of the most important Cu–Au and Au–PGE provinces globally, with deposits extending along an approximately 150 km long WNW-trending zone about 60 km wide centred on the Carajás Fault. The larger deposits (approx. 200–1000 Mt @ 0.95–1.4% Cu and 0.3–0.85 g/t Au) are classic Fe-oxide Cu–Au deposits that include Salobo, Igarapé Bahia–Alemão, Cristalino and Sossego. They are largely hosted in the lower volcanic sequences and basement gneisses as pipe- or ring-like mineralised, generally breccia bodies that are strongly Fe- and LREE-enriched, commonly with anomalous Co and U, and quartz- and sulfur-deficient. Iron oxides and Fe-rich carbonates and/or silicates are invariably present. Rhenium–Os dating of molybdenite at Salobo and SHRIMP Pb–Pb dating of hydrothermal monazite at Igarapé-Bahia indicate ages of ca. 2.57 Ga for mineralisation, indistinguishable from ages of poorly-exposed Archean alkalic and A-type intrusions in the Itacaiúnas Belt, strongly implicating a deep magmatic connection.A group of smaller, commonly supergene-enriched Cu–Au deposits (generally < 50 Mt @ < 2% Cu and < 1 g/t Au in hypogene ore), with enrichment in granitophile elements such as W, Sn and Bi, spatially overlap the Archean Fe-oxide Cu–Au deposits. These include the Breves, Águas Claras, Gameleira and Estrela deposits which are largely hosted by the upper sedimentary sequence as greisen-to ring-like or stockwork bodies. They generally lack abundant Fe-oxides, are quartz-bearing and contain more S-rich Cu–Fe sulfides than the Fe-oxide Cu–Au deposits, although Cento e Dezoito (118) appears to be a transitional type of deposit. Precise Pb–Pb in hydrothermal phosphate dating of the Breves and Cento e Dezoito deposits indicate ages of 1872 ± 7 Ma and 1868 ± 7 Ma, respectively, indistinguishable from Pb–Pb ages of zircons from adjacent A-type granites and associated dykes which range from 1874 ± 2 Ma to 1883 ± 2 Ma, with 1878 ± 8 Ma the age of intrusions at Breves. An unpublished Ar/Ar age for hydrothermal biotite at Estrela is indistinguishable, and a Sm–Nd isochron age for Gameleira is also similar, although somewhat younger. The geochronological data, combined with geological constraints and ore-element associations, strongly implicate a magmatic connection for these deposits.The highly anomalous, hydrothermal Serra Pelada Au–PGE deposit lies at the north-eastern edge of the Province within the same fault corridor as the Archean and Paleoproterozoic Cu–Au deposits, and like the Cu–Au deposits is LREE enriched. It appears to have formed from highly oxidising ore fluids that were neutralised by dolomites and reduced by carbonaceous shales in the upper sedimentary succession within the hinge of a reclined synform. The imprecise Pb–Pb in hydrothermal phosphate age of 1861 ± 45 Ma, combined with an Ar/Ar age of hydrothermal biotite of 1882 ± 3 Ma, are indistinguishable from a Pb–Pb in zircon age of 1883 ± 2 Ma for the adjacent Cigano A-type granite and indistinguishable from the age of the Paleoproterozoic Cu–Au deposits. Again a magmatic connection is indicated, particularly as there is no other credible heat or fluid source at that time.Finally, there is minor Au–(Cu) mineralisation associated with the Formiga Granite whose age is probably ca. 600 Ma, although there is little new zircon growth during crystallisation of the granite. This granite is probably related to the adjacent Neoproterozoic (900–600 Ma) Araguaia Fold Belt, formed as part of the Brasiliano Orogeny.Thus, there are two major and one minor period of Cu–Au mineralisation in the Carajás Mineral Province. The two major events display strong REE enrichment and strongly enhanced LREE. There is a trend from strongly Fe-rich, low-SiO₂ and low-S deposits to quartz-bearing and more S-rich systems with time. There cannot be significant connate or basinal fluid (commonly invoked in the genesis of Fe-oxide Cu–Au deposits) involved as all host rocks were metamorphosed well before mineralisation: some host rocks are at mid- to high-amphibolite facies. The two major periods of mineralisation correspond to two periods of alkalic to A-type magmatism at ca. 2.57 Ga and ca. 1.88 Ga, and a magmatic association is compelling.The giant to world-class late Archean Fe-oxide Cu–Au deposits show the least obvious association with deep-seated alkaline bodies as shown at Palabora, South Africa, and implied at Olympic Dam, South Australia. The smaller Paleoproterozoic Cu–Au–W–Sn–Bi deposits and Au–PGE deposit show a more obvious relationship to more fractionated A-type granites, and the Neoproterozoic Au–(Cu) deposit to crustally-derived magmas. The available data suggest that magmas and ore fluids were derived from long-lived metasomatised lithosphere and lower crust beneath the eastern margin of the Amazon Craton in a tectonic setting similar to that of other large Precambrian Fe-oxide Cu–Au deposits.     

Methane-derived authigenic carbonates from the northern Gulf of Mexico — MD02 Cruise

Authigenic carbonates were sampled in piston cores collected from both the Tunica Mound and the Mississippi Canyon area on the continental slope of the northern Gulf of Mexico during a Marion Dufresne cruise in July 2002. The carbonates are present as hardgrounds, porous crusts, concretions or nodules and shell fragments with or without carbonate cements. Carbonates occurred at gas venting sites which are likely to overlie gas hydrates bearing sediments. Electron microprobe, X-ray diffraction (XRD) and thinsection investigations show that these carbonates are high-Mg calcite (6–21 mol% MgCO₃), with significant presence of framboidal pyrite. All carbonates are depleted in ¹³C (δ¹³C = − 61.9 to − 31.5‰ PDB) indicating that the carbon is derived mainly from anaerobic methane oxidation (AMO). Age estimates based on ¹⁴C dating of shell fragments and on regional sedimentation rates indicate that these authigenic carbonates formed within the last 1000 yr in the Mississippi Canyon and within 5500 yr at the Tunica Mound. The oxygen isotopic composition of carbonates ranges from + 3.4 to + 5.9‰ PDB. Oxygen isotopic compositions and Mg²⁺ contents of carbonates, and present in-situ temperatures of bottom seawater/sediments, show that some of these carbonates, especially from a core associated with underlying massive gas hydrates precipitated in or near equilibrium with bottom-water. On the other hand, those carbonates more enriched in ¹⁸O are interpreted to have precipitated from ¹⁸O-rich fluids which are thought to have been derived from the dissociation of gas hydrates. The dissociation of gas hydrates in the northern Gulf of Mexico within the last 5500 yr may be caused by nearby salt movement and related brines.     

An early Paleoproterozoic high-K intrusive complex in southwestern Tarim Block, NW China: Age, geochemistry, and tectonic implications

Systematic geochronologic, geochemical, and Nd isotopic analyses were carried out for an early Paleoproterozoic high-K intrusive complex exposed in southwestern Tarim, NW China. The results provide a better understanding of the Paleoproterozoic tectonic evolution of the Tarim Block. Zircon U–Pb age dating indicates two Paleoproterozoic magmatic episodes occurring at ca. 2.41 Ga and ca. 2.34 Ga respectively, which were followed by a ca. 1.9 Ga metamorphic event. The 2.41 Ga granodiorite–adamellite suite shares characteristics of late to post-orogenic metaluminous A-type granites in its high alkalinity (Na₂O + K₂O = 7.6–9.3%), total REE (410–788 ppm), Zr (370–660 ppm), and Y (21.7–58.4 ppm) contents. εNd(t) values for the suite range from − 3.22 to − 4.71 and accordingly the Nd modal ages (T_2DM) vary between 3.05 Ga and 3.17 Ga. Based on geochemical data, the 2.34 Ga suite can be subdivided into two sub-suites, namely A-type and S-type. However, both types have comparable Nd isotope compositions (εNd(t) ≈ − 0.41 to − 2.08) and similar narrow T_2DM ranges (2.76–2.91 Ga).Geochemical and Nd isotopic data for the high-K intrusive complex, in conjunction with the regional geological setting, suggest that both the 2.41 Ga suite and the 2.34 Ga A-type sub-suite might have been produced by partial melting of the Archean mafic crust in a continental rift environment. The S-type sub-suite is thought to have formed by partial melting of felsic pelites and/or metagreywackes recycled from Archean crust (TTG?). Gabbro enclaves with positive εNd(t) value (2.15) have been found to be intermingling within the 2.34 Ga suite; ca. 2.34–2.36 Ga gabbroic dykes and adamellites have previously been documented in eastern Tarim. These observations indicate that the high-K intrusions may reflect the emergence of depleted mantle upwelling beneath the Tarim Block at that time. We suggest a three-stages model for the Precambrian crustal evolution in the Tarim Block: (1) the formation of proto-crust (TTG) by ca. 2.5 Ga, (2) episodes of felsic magmatism possibly occurring in continental rift environments at ca. 2.41 Ga and ca. 2.34–2.36 Ga, and (3) ca. 1.9 Ga metamorphism that may represent the solidification of the Precambrian basement of the Tarim Block.     

Neoarchaean high-grade metamorphism in the Central Zone of the Limpopo Belt (South Africa): Combined petrological and geochronological evidence from the Bulai pluton

The Bulai pluton represents a calc-alkaline magmatic complex of variable deformed charnockites, enderbites and granites, and contains xenoliths of highly deformed metamorphic country rocks. Petrological investigations show that these xenoliths underwent a high-grade metamorphic overprint at peak P–T conditions of 830–860 °C/8–9 kbar followed by a pressure–temperature decrease to 750 °C/5–6 kbar. This P–T path is inferred from the application of P–T pseudosections to six rock samples of distinct bulk composition: three metapelitic garnet–biotite–sillimanite–cordierite–plagioclase–(K-feldspar)–quartz gneisses, two charnoenderbitic garnet–orthopyroxene–biotite–K-feldspar–plagioclase–quartz gneisses and an enderbitic orthopyroxene–biotite–plagioclase–quartz gneiss. The petrological data show that the metapelitic and charnoenderbitic gneisses underwent uplift, cooling and deformation before they were intruded by the Bulai Granite. This relationship is supported by geochronological results obtained by in situ LA-ICP-MS age dating. U–Pb analyses of monazite enclosed in garnet of a charnoenderbite gneiss provide evidence for a high-grade structural-metamorphic–magmatic event at 2644 ± 8 Ma. This age is significantly older than an U–Pb zircon crystallisation age of 2612 ± 7 Ma previously obtained from the surrounding, late-tectonic Bulai Granite. The new dataset indicates that parts of the Limpopo's Central Zone were affected by a Neoarchaean high-grade metamorphic overprint, which was caused by magmatic heat transfer into the lower crust in a ‘dynamic regional contact metamorphic milieu’, which perhaps took place in a magmatic arc setting.     

Two magma series and associated ore deposit types in the Permian Emeishan large igneous province, SW China

The Late Middle Permian ( 260 Ma) Emeishan large igneous province in SW China contains two magmatic series, one comprising high-Ti basalts and Fe-rich gabbroic and syenitic intrusions, the other low-Ti basalts and mafic–ultramafic intrusions. The Fe-rich gabbros are spatially and temporally associated with syenites. Each series is associated with a distinctive type of mineralization, the first with giant Fe–Ti–V oxide ore deposits such as Panzhihua and Baima, the second with Ni–Cu–(PGE) sulfide deposits such as Jinbaoshan, Limahe and Zhubu. New SHRIMP zircon U–Pb isotopic data yielded 263 ± 3 Ma for the Limahe intrusion, 261 ± 2 Ma for the Zhubu intrusion and 262 ± 2 Ma for a syenitic intrusion. These new age dates, together with previously reported SHRIMP zircon U–Pb ages, suggest that all these intrusions are contemporaneous with the Emeishan flood basalts and formed during a major igneous event at ca. 260 Ma.The oxide-bearing intrusions have higher Al₂O₃, FeO (as total iron) and total alkalis (Na₂O + K₂O) but lower MgO than the sulfide-bearing intrusions. All intrusions are variably enriched in LREE relative to HREE. The oxide-bearing intrusions display positive Nb- and Ti-anomalies and in certain cases negative Zr–Hf anomalies, whereas the sulfide-bearing intrusions have obvious negative Nb- and Ti-anomalies, a feature of crustal contamination. Individual intrusions have relatively small ranges of Nd(t) values. All the intrusions, however, have Nd(t) values ranging from − 3.9 to + 4.6, and initial ⁸⁷Sr/⁸⁶Sr ratios from 0.7039 to 0.7105. The syenites have very low MgO (< 2 wt.%) but highly variable Fe₂O₃ (2.5 to 13 wt.%) with initial ⁸⁷Sr/⁸⁶Sr ratios ranging from 0.7039 to 0.7089. Magmas from both series could have derived by melting of a heterogeneous mantle plume: the high-Ti series from a Fe-rich, more fertile source and the low-Ti series from a Fe-poor, more refractory source. In addition, the low-Ti series underwent significant crustal contamination. The two magma series evolved along different paths that led to distinct mineralization styles.     

A palaeomagnetic study of charnockites from Madras Block, Southern Granulite Terrain, India

The South Indian Craton is composed of low-grade and high-grade metamorphic rocks across different tectonic blocks between the Moyar–Bhavani and Palghat–Cauvery shear zones and an elongated belt of eastern margin of the peninsular shield. The Madras Block north of the Moyar–Bhavani shear zone, which evolved throughout the Precambrian period, mainly consists of high-grade metamorphic rocks. In order to constrain the evolution of the charnockitic region of the Pallavaram area in the Madras Block we have undertaken palaeomagnetic investigation at 12 sites. ChRM directions in 61 oriented block samples were investigated by Alternating Field (AF) and Thermal demagnetization. Titanomagnetite in Cation Deficient (CD) and Multi Domain (MD) states is the remanence carrier. The samples exhibit a ChRM with reverse magnetization of D_m = 148.1, I_m = + 48.6 (K = 22.2, α₉₅ = 9.0) and a palaeomagnetic pole at 37.5 °N, 295.6 °E (dp/dm = 7.8°/11.8°). This pole plots at a late Archaean location on the Indian Apparent Polar Wander Path (APWP) suggesting an age of magnetization in the Pallavaram charnockites as 2600 Ma. The nearby St. Thomas Mount charnockites indicate a period of emplacement at 1650 Ma (Mesoproterozoic). Thus the results of Madras Block granulites also reveal crustal evolution similar to those in the Eastern Ghats Belt with identical palaeopoles from both the areas.     

Pressure–temperature evolution of lawsonite eclogite in Sivrihisar; Tavşanlı Zone–Turkey

The Cretaceous blueschist belt, Tavşanlı Zone, representing the subducted and exhumed northern continental margin of the Anatolide–Tauride platform is exposed in Western Anatolia. The Sivrihisar area east of Tavşanlı is made up of tectonic units consisting of i) metaclastics and conformably overlying massive marbles (coherent blueschist unit), ii) blueschist-eclogite unit, iii) marble–calcschist intercalation and iv) metaperidotite slab. The metaclastics are composed of jadeite–lawsonite–glaucophane and jadeite–glaucophane–chloritoid schists, phengite phyllites, and calcschists with glaucophane–lawsonite metabasite layers. The blueschist-eclogite unit representing strongly sheared, deeply buried and imbricated tectonic slices of accreted uppermost levels of the oceanic crust with minor metamorphosed serpentinite bodies consists of lawsonite-bearing eclogitic metabasites (approximately 90% of the field), lawsonite eclogites, metagabbros, serpentinites, pelagic marbles, omphacite–glaucophane–lawsonite metapelites and metacherts. The mineral assemblage of the lawsonite eclogite (garnet + omphacite > 70%) is omphacite, garnet, lawsonite, glaucophane, phengite and rutile. Lawsonite eclogite lenses are enclosed by garnet–lawsonite blueschist envelopes.Textural evidence from lawsonite eclogites and country rocks reveals that they did not leave the stability field of lawsonite during subduction and exhumation. The widespread preservation of lawsonite in eclogitic metabasites and eclogites can be attributed to rapid subduction and subsequent exhumation in a low geothermal gradient of the oceanic crust material without experiencing a thermal relaxation. Peak P–T conditions of lawsonite eclogites are estimated at 24 ± 1 kbar and 460 ± 25 °C. These P–T conditions indicate a remarkably low geotherm of 6.2 °C/km corresponding to a burial depth of 74 km.     

Magnetic fabrics and rock magnetism of Archaean and Proterozoic dike swarms in the southern São Francisco Craton, Brazil

Magnetic fabric and rock magnetism studies were performed on three mafic dike swarms (total of 38 dikes) from the southernmost part of the São Francisco Craton (SFC) (Minas Gerais State, SE Brazil). They cut Archaean granite–gneiss–migmatite and paleoprototerozoic terranes. These swarms are classified as basic–noritic (Sm–Nd age  2.65 Ga), basic (Rb–Sr age  1.87 Ga) and metamorphic (Rb–Sr age  1.87 Ga) suites, in which the second is the most important. Magnetic fabrics were determined by applying both anisotropy of low-field magnetic susceptibility (AMS) and anisotropy of anhysteretic remanent magnetization (AARM). In most sites magnetic susceptibility is dominantly carried by ferromagnetic minerals, however, in some sites the paramagnetic contribution exceeds 70% of bulk susceptibility. Mainly coarse to fine-grained Ti-poor titanomagnetite up to pure magnetite carry the magnetic fabrics.Three primary AMS fabrics are recognized which are all coaxial with the AARM fabric. Normal AMS fabric is dominant in the basic suite (16 of 20 analyzed dikes) and occurs in 4 and 3 dikes from the basic–noritic and metamorphic suites, respectively. This fabric is interpreted as a result of magma flow in which the analysis of K_max inclination permitted to infer that the majority of dikes were fed by inclined flows (30° < K_max < 60°), although 44% of dikes from the basic suite were fed by horizontal or sub-horizontal flows (K_max < 30°). Intermediate AMS fabric was found in 50% of dikes from the basic–noritic and metamorphic suites, but in only 2 dikes from the basic suite. It is interpreted as due to vertical compaction of a static magma column with the minimum stress along the dike strike. Inverse AMS fabric is a minority (2 dikes from each suite). The parallelism between AMS and AARM tensors for dikes with abnormal fabrics suggests a primary origin for them. Gyroremanent magnetization (GRM) effect was negligible for the majority of dikes, but it was found in two dikes from the basic suite with normal AMS fabric.Magnetic fabrics recognized for the three studied swarms do not depend on magnetic mineralogy, geochemical composition, dike strikes, nor the age of the swarms since the same magnetic minerals and magnetic fabric types are found in dikes from all suites. Inclined and horizontal flows allow us to infer the relative position of at least three magma sources (or magma chambers) from which the dikes were fed.     

Draa Sfar, Morocco: A Visean (331 Ma) pyrrhotite-rich, polymetallic volcanogenic massive sulphide deposit in a Hercynian sediment-dominant terrane

Draa Sfar is a Visean, stratabound, volcanogenic massive sulphide ore deposit hosted by a Hercynian carbonaceous, black shale-rich succession of the Jebilet terrane, Morocco. The ore deposit contains 10 Mt grading 5.3 wt.% Zn, 2 wt.% Pb, and 0.3 wt.% Cu within two main massive sulphides orebodies, Tazakourt (Zn-rich) and Sidi M'Barek (Zn–Cu rich). Pyrrhotite is by far the dominant sulphide (70 to 95% of total sulphides), sphalerite is fairly abundant, chalcopyrite and galena are accessory, pyrite, arsenopyrite and bismuth minerals are rare. Pyrrhotite is monoclinic and mineralogical criteria indicate that it is of primary origin and not formed during metamorphism. Its composition is very homogeneous, close to Fe₇S₈, and its absolute magnetic susceptibility is 2.10^− 3 SI/g. Ar–Ar dating of hydrothermal sericites from a coherent rhyolite flow or dome within the immediate deposit footwall indicates an age of 331.7 ± 7.9 Ma for the Draa Sfar deposit and rhyolite volcanism.The Draa Sfar deposit has undergone a low-grade regional metamorphic event that caused pervasive recrystallization, followed by a ductile–brittle deformation event that has locally imparted a mylonitic texture to the sulphides and, in part, is responsible for the elongated and sheet-like morphology of the sulphide orebodies. Lead isotope data fall into two compositional end-members. The least radiogenic end-member, (²⁰⁶Pb/²⁰⁴Pb = 18.28), is characteristic of the Tazakourt orebody, whereas the more radiogenic end-member (²⁰⁶Pb/²⁰⁴Pb  18.80) is associated with the Sidi M'Barek orebody, giving a mixing trend between the two end-members. Lead isotope compositions at Draa Sfar testify to a significant continental crust source for the base metals, but are different than those of the Hajar and South Iberian Pyrite Belt VMS deposits.The abundance of pyrrhotite versus pyrite in the orebodies is attributed to low fO₂ conditions and neither a high temperature nor a low aH₂S (below 10^− 3) is required. The highly anoxic conditions required to stabilize pyrrhotite over pyrite are consistent with formation of the deposit within a restricted, sediment-starved, anoxic basin characterized by the deposition of carbonaceous, pelagic sediments along the flank of a rhyolitic flow-dome complex that was buried by pelitic sediments. Deposition of sulphides likely occurred at and below the seafloor within anoxic and carbonaceous muds.Draa Sfar and other Moroccan volcanogenic massive sulphide deposits occur in an epicontinental volcanic domain within the outer zone of the Hercynian belt and formed within a sedimentary environment that has a high pelagic component. In spite of the diachronous emplacement between the IPB deposits (late Devonian to Visean) and Moroccan deposits (Dinantian), all were formed around 340 ± 10 Ma following a major phase of the Devonian compression.     

Geochronological framework and Pb, Sr isotope geochemistry of the Qingchengzi Pb–Zn–Ag–Au orefield, Northeastern China

The Qingchengzi orefield in northeastern China, is a concentration of several Pb–Zn, Ag, and Au ore deposits. A combination of geochronological and Pb, Sr isotopic investigations was conducted. Zircon SHRIMP U–Pb ages of 225.3 ± 1.8 Ma and 184.5 ± 1.6 Ma were obtained for the Xinling and Yaojiagou granites, respectively. By step-dissolution Rb–Sr dating, ages of 221 ± 12 Ma and 138.7 ± 4.1 Ma were obtained for the sphalerite of the Zhenzigou Zn–Pb deposit and pyrargyrite of the Ag ore in the Gaojiabaozi Ag deposit, respectively. Pb isotopic ratios of the Ag ore at Gaojiabaozi (²⁰⁶Pb/²⁰⁴Pb = 18.38 to 18.53) are higher than those of the Pb–Zn ores (²⁰⁶Pb/²⁰⁴Pb = 17.66 to 17.96; Chen et al. [Chen, J.F., Yu, G., Xue, C.J., Qian, H., He, J.F., Xing, Z., Zhang, X., 2005. Pb isotope geochemistry of lead, zinc, gold and silver deposit clustered region, Liaodong rift zone, northeastern China. Science in China Series D 48, 467–476.]). Triassic granites show low Pb isotopic ratios (²⁰⁶Pb/²⁰⁴Pb = 17.12 to 17.41, ²⁰⁷Pb/²⁰⁴Pb = 15.47 to 15.54, ²⁰⁸Pb/²⁰⁴Pb = 37.51 to 37.89) and metamorphic rocks of the Liaohe Group have high ratios (²⁰⁶Pb/²⁰⁴Pb = 18.20 to 24.28 and 18.32 to 20.06, ²⁰⁷Pb/²⁰⁴Pb = 15.69 to 16.44 and 15.66 to 15.98, ²⁰⁸Pb/²⁰⁴Pb = 37.29 to 38.61 and 38.69 to 40.00 for the marble of the Dashiqiao Formation and schist of the Gaixian Formation, respectively).Magmatic activities at Qingchengzi and in adjacent regions took place in three stages, and each contained several magmatic pulses: ca. 220 to 225 Ma and 211 to 216 Ma in the Triassic; 179 to 185 Ma, 163 to 168 Ma, 155 Ma and 149 Ma in the Jurassic, as well as ca. 140 to 130 Ma in the Early Cretaceous. The Triassic magmatism was part of the Triassic magmatic belt along the northern margin of the North China Craton produced in a post-collisional extensional setting, and granites in it formed by crustal melting induced by mantle magma. The Jurassic and Early Cretaceous magmatism was related to the lithospheric delamination in eastern China. The Triassic is the most important metallogenic stage at Qingchengzi. The Pb–Zn deposits, the Pb–Zn–Ag ore at Gaojiabaozi, and the gold deposits were all formed in this stage. They are temporally and spatially associated with the Triassic magmatic activity. Mineralization is very weak in the Jurassic. Ag ore at Gaojiabaozi was formed in the Early Cretaceous, which is suggested by the young Rb–Sr isochron age, field relations, and significantly different Pb isotopic ratios between the Pb–Zn–Ag and Ag ores. Pb isotopic compositions of the Pb–Zn ores suggest binary mixing for the source of the deposits. The magmatic end-member is the Triassic granites and the other metamorphic rocks of the Liaohe Group. Slightly different proportions of the two end-members, or an involvement of materials from hidden Cretaceous granites with slightly different Pb isotopic ratios, is postulated to interpret the difference of Pb isotopic compositions between the Pb–Zn–(Ag) and Ag ores. Sr isotopic ratios support this conclusion. At the western part of the Qingchengzi orefield, hydrothermal fluid driven by the heat provided by the now exposed Triassic granites deposited ore-forming materials in the low and middle horizons of the marbles of the Dashiqiao Formation near the intrusions to form mesothermal Zn–Pb deposits. In the eastern part, hydrothermal fluids associated with deep, hidden Triassic intrusions moved upward along a regional fault over a long distance and then deposited the ore-forming materials to form epithermal Au and Pb–Zn–Ag ores. Young magmatic activities are all represented by dykes across the entire orefield, suggesting that the corresponding main intrusion bodies are situated in the deep part of the crust. Among these, only intrusions with age of ca. 140 Ma might have released sufficient amounts of fluid to be responsible for the formation of the Ag ore at Gaojiabaozi.Our age results support previous conclusions that sphalerite can provide a reliable Rb–Sr age as long as the fluid inclusion phase is effectively separated from the “sulfide” phase. Our work suggests that the separation can be achieved by a step-resolution technique. Moreover, we suggest that pyrargyrite is a promising mineral for Rb–Sr isochron dating.     

Metamorphic path constrained by metapelitic rocks from the inner Aiuruoca-Andrelândia nappe, south of the São Francisco craton, SE Brazil

A Nappe system south to southwest of the São Francisco Craton represents the southern extension of the Brasília belt and describes an inverted metamorphic pile of greenschist facies toward amphibolite facies. The Aiuruoca-Andrelândia nappe is one of the nappes of this system. The hind portion of the Aiuruoca-Andrelândia nappe, south of Caxambu and Aiuruoca (MG), consists of a structural-metamorphic domain transported toward the E-NE. There is a metamorphic transition, from the kyanite zone to kyanite and sillimanite coexistence, until the sillimanite zone. Metapelitic rocks preserve high-pressure parageneses (Rt–Ky–Grt–Ms–Bt–Pl–Qtz) and contain retrograde eclogitic rocks. Sil–Pl–Qtz coronitic intergrowths around garnets are common decompressive textures. Kyanite schists register the P_max of 11 kbar at 660 °C and define a decompressive path until 6–7.5 kbar at 650 °C. These P–T conditions represent the equilibrium in S₂ schistosity (amphibolite facies) and the beginning of the cooling path in the Ky–Sil transition. The decompressive path suggests an extrusional process, immediately after burying at about 60 km. Exhumation controlled by convergent events, related to the São Francisco Plate subduction and tectonic erosion, took these units, isothermally, to higher levels (20–33 km). Later, the metamorphic path shifted toward near-isobaric cooling.     

The subducted oceanic crust within continental-type UHP metamorphic belt in the North Qaidam, NW China: Evidence from petrology, geochemistry and geochronology

Three types of eclogite, together with a serpentinized harzburgite, coexist as blocks within granitic and pelitic gneisses along the Shaliuhe cross section, the eastern part of the North Qaidam continental-type ultrahigh-pressure (UHP) metamorphic belt, NW China. The olivine (Ol¹) and orthopyroxene in the harzburgite are compositionally similar to present-day abyssal peridotites. The kyanite–eclogite is derived from a troctolitic protolith, whereas the epidote–eclogite from a gabbroic protolith, both having distinct positive Eu anomalies, low TiO₂, and high Al₂O₃ and MgO. The kyanite–eclogite shows inherited cumulate layering. The phengite–eclogite has high TiO₂, low Al₂O₃ and MgO with incompatible trace elements resembling enriched-type MORB. Sr–Nd isotope data indicate that the protoliths of both kyanite–eclogite and epidote–eclogite ([⁸⁷Sr/⁸⁶Sr]_i ~ 0.703–0.704; ε_Nd(T) ~ 5.9–8.0) are of mantle origin (e.g., ocean crust signatures). On the other hand, while the lower ε_Nd(T) value (1.4–4.1) of phengite–eclogite is more or less consistent with an enriched MORB protolith, their high [⁸⁷Sr/⁸⁶Sr]_i ratio (0.705–0.716) points to an additional enrichment in their history, probably in an subduction-zone environment. Field relations and geochemical analyses suggest that the serpentinized harzburgite and the three types of eclogite constitute the oceanic lithological section of an ophiolitic sequence from mantle peridotite, to cumulate, and to upper basaltic rocks. The presence of coesite pseudomorphs and quartz exsolution in omphacite plus thermobarometric calculations suggests that the eclogites have undergone ultrahigh pressure metamorphism (i.e., peak P ≥ 2.7 GPa). The harzburgite may also have experienced the same metamorphism, but the lack of garnet suggests that the pressure conditions of ≤ 3.0 GPa. Zircon U–Pb SHRIMP dating shows that the eclogites have a protolith age of 516 ± 8 Ma and a metamorphic age of 445 ± 7 Ma. These data indicate the presence of a Paleo-Qilian Ocean between Qaidam and Qilian blocks before the early Ordovician. The ophiolitic assemblage may be the relics of subducted oceanic crust prior to the subduction of continental materials during Ordovician–Silurian times and ultimate continent collision. These rocks, altogether, record a complete history of ocean crust subduction, to continental subduction, and to continental collision.     

The Archean BIF-hosted Cuiabá Gold deposit, Quadrilátero Ferrífero, Minas Gerais, Brazil

The Cuiabá Gold Deposit is located in the northern part of the Quadrilátero Ferrífero, Minas Gerais State, Brazil. The region constitutes an Archean granite–greenstone terrane composed of a basement complex (ca. 3.2 Ga), the Rio das Velhas Supergroup greenstone sequence, and related granitoids (3.0–2.7 Ga), which are overlain by the Proterozoic supracrustal sequences of the Minas (< 2.6–2.1  Ga) and Espinhaço (1.7 Ga) supergroups.The stratigraphy of the Cuiabá area is part of the Nova Lima Group, which forms the lower part of the Rio das Velhas Supergroup. The lithological succession of the mine area comprises, from bottom to top, lower mafic metavolcanics intercalated with carbonaceous metasedimentary rocks, the gold-bearing Cuiabá-Banded Iron Formation (BIF), upper mafic metavolcanics and volcanoclastics and metasedimentary rocks. The metamorphism reached the greenschist facies. Tectonic structures of the deposit area are genetically related to deformation phases D₁, D₂, D₃, which took place under crustal compression representing one progressive deformational event (E_n).The bulk of the economic-grade gold mineralization is related to six main ore shoots, contained within the Cuiabá BIF horizon, which range in thickness between 1 and 6 m. The BIF-hosted gold orebodies (> 4 ppm Au) represent sulfide-rich segments of the Cuiabá BIF, which grade laterally into non-economic mineralized or barren iron formation. Transitions from sulfide-rich to sulfide-poor BIF are indicated by decreasing gold grades from over 60 ppm to values below the fire assay detection limit in sulfide-poor portions. The deposit is “gold-only”, and shows a characteristic association of Au with Ag, As, Sb and low base-metal contents. The gold is fine grained (up to 60 μm), and is generally associated with sulfide layers, occurring as inclusions, in fractures or along grain boundaries of pyrite, the predominant sulfide mineral (> 90 vol.%). Gold is characterized by an average fineness of 0.840 and a large range of fineness (0.759 to 0.941).The country rocks to the mineralized BIF show strong sericite, carbonate and chlorite alteration, typical of greenschist facies metamorphic conditions. Textures observed on microscopic to mine scales indicate that the mineralized Cuiabá BIF is the result of sulfidation involving pervasive replacement of Fe-carbonates (siderite–ankerite) by Fe-sulfides. Gold mineralization at Cuiabá shows various features reported for Archean gold–lode deposits including the: (1) association of gold mineralization with Fe-rich host rocks; (2) strong structural control of the gold orebodies, showing remarkable down-plunge continuity (> 3 km) relative to strike length and width (up to 20 m); (3) epigenetic nature of the mineralization, with sulfidation as the major wall–rock alteration and directly associated with gold deposition; (4) geochemical signature, with mineralization showing consistent metal associations (Au–Ag–As–Sb and low base metal), which is compatible with metamorphic fluids.     

Quantitative reconstruction of Holocene precipitation changes in southern Patagonia

Holocene variations in annual precipitation (P_ann) were reconstructed from pollen data from southern Argentinian Patagonia using a transfer function developed based on a weighted-averaging partial least squares (WA-PLS) regression. The pollen–climate calibration model consisted of 112 surface soil samples and 59 pollen types from the main vegetation units, and modern precipitation values obtained from a global climate database. The performance (r² = 0.517; RMSEP = 126 mm) of the model was comparable or slightly lower than in other comparable pollen–climate models. Fossil pollen data were obtained from a sediment core from Cerro Frias site (50°24'S, 72°42'W) located at the forest-steppe ecotone. Reconstructed P_ann values of about 200 mm suggest dry conditions during the Pleistocene–Holocene transition (12,500–10,500 cal yr BP). P_ann values were about 300–350 mm from 10,500 to 8000 cal yr BP and increased to 400–500 mm between 8000 and 1000 cal yr BP. An abrupt decrease in P_ann at about 1000 cal yr BP was associated with a Nothofagus decline. The reconstructed P_ann suggests a weakening and southward shift of the westerlies during the early Holocene and intensification, with no major latitudinal shifts, during the mid-Holocene at high latitudes in southern Patagonia.