S. Hoernes

Petrology and geochemistry of syn- to post-collisional metaluminous A-type granites - a major and trace element and Nd-Sr-Pb-O-isotope study from the Proterozoic Damara Belt, Namibia

Abstract Leucocratic A-type granites of the Proterozoic Damara orogen, Namibia are a distinctive groups of rocks based on their mineralogical, chemical and isotopic characteristics. A-type granites can be distinguished from the leucocratic S-type granites on the basis of Fe-rich amphibole and biotite and higher abundances of Zr, Nb, Y and LREE but lower values of 87 Sr / 86 Sr (500 Ma ) and δ 18 O in the A-type granites. In comparison with A-type granites described in the literature, the Damaran A-type granites are enriched in CaO, Al2O3, MgO, FeO, Ba, Sr and Eu but depleted in Ga and Zn. Low K/Rb, Rb/Sr, Rb/Ba and Sr/Ba ratios indicate that these melts are not fractionation products of calc-alkaline source rocks and therefore, these chemical criteria are interpreted as a feature of the source rocks, i.e., undepleted metatonalites. The A-type granites have high calculated zircon and monazite saturation temperatures mostly in excess of 800°C and it is concluded that these granites represent high-temperature partial melts. The relationship between major- and trace element data and isotopic composition demonstrates that the A-type granites originate from several sources and have undergone different processes. The lack of evidence for shallow crustal contamination (e.g., lack of country rock xenoliths), the low viscosities of the melts (as inferred from their high calculated temperature) and their high Nd concentrations relative to the country rocks suggest that the isotopic characteristics were acquired from an enriched mantle. Initial Sr isotopic compositions vary over a narrow range (0.70486–0.70928) but the initial Nd isotopic compositions vary considerably with initial eNd values between −0.03 and −6.35. This wide range in the Nd isotope values is probably the result of interaction of mantle-derived magmas with crustal rocks. A-type granites at Baukwab have negatively correlated initial eNd values and 207 Pb / 204 Pb and 206 Pb / 204 Pb ratios indicating interaction of mantle-derived magmas (low initial eNd, low 207 Pb / 204 Pb and 206 Pb / 204 Pb ) with continental crust (high initial eNd, high 207 Pb / 204 Pb and 206 Pb / 204 Pb ). A-type granites from Albrechtstal and Oetmoed have positively correlated Pb isotope ratios and show also positive correlations between Nd isotopes and Pb isotopes, suggesting interaction of at least two different source materials. The relationship between Nd isotopes and Nd concentrations precludes simple assimilation, instead mixing of different sources seems to be more realistic. For the A-type granites from Oetmoed, oxygen isotopes are negatively correlated with Sr isotopes indicating that the crustal end-member has high δ 18 O values and rather low initial 87 Sr / 86 Sr ratios whereas a hypothetical mantle component has lower δ 18 O values but higher initial 87 Sr / 86 Sr ratios. The chemical and isotope data suggest that the Damaran A-type granites were generated by a variety of processes including partial melting of mantle-derived tonalitic sources, limited crystal fractionation and interaction of these magmas with crustal material. These data indicate that the Damaran A-type suites mark episodes of continental growth during which mantle material, although limited in extent, is added to the crust.

Geochronology and petrogenesis of Pan-African, syn-tectonic, S-type and post-tectonic A-type granite (Namibia): products of melting of crustal sources, fractional crystallization and wall rock entrainment

Abstract The Oetmoed Granite–Migmatite Complex (OGMC), Central Damara Orogen, Namibia, consists mainly of ∼526 to ∼516 Ma garnet- and cordierite-bearing granite and subordinate ∼488 to ∼494 Ma hornblende- and titanite-bearing granite in the form of planar sheets and dykes. Additionally, a slightly elongated granite body occurs in the center of the complex. The garnet- and cordierite-bearing granite has major- and trace-element characteristics of S-type granite but the hornblende- and titanite-bearing granite has higher HFSE and REE contents similar to A-type granite. Whereas the garnet- and cordierite-bearing granite contains numerous restitic xenoliths, the hornblende- and titanite-bearing granite is xenolith-free. The country rocks are cordierite–sillimanite–K-feldspar–garnet-bearing metasedimentary rocks and migmatite. Cordierite- and garnet-rich xenoliths in the S-type granite do not represent primary restite, their depleted chemical composition is best explained by varying and large degrees of partial melting of incorporated country rocks. Most chemical variations among the garnet- and cordierite-bearing granite can be explained by processes linked with fractional crystallization of plagioclase, biotite and accessory phases, mostly monazite and zircon. Major and trace element data and high δ 18 O values suggest that the least evolved members of the garnet- and cordierite-bearing granite were derived from metapelitic rocks at ca. 800°C as inferred from monazite and apatite dissolution thermometry. Higher CaO and Na2O but lower SiO2 contents and lower Rb/Sr ratios as well as lower δ 18 O values of the hornblende- and titanite-bearing granite suggest that they are more likely generated by partial melting of non-pelitic sources (metagranitoids?) at temperatures in excess of 900°C. Decreasing TiO2, Na2O, FeOtot., MgO, CaO, total REE content but increasing Al2O3 and K2O indicate fractionation of mainly hornblende and titanite in the case of the hornblende- and titanite-bearing granite. The differing compositions of the garnet- and cordierite-bearing granite and the hornblende- and titanite-bearing granite are attributed to different source rocks (metapelite instead of metagranitoid) and different temperatures during melting as inferred from accessory phase dissolution thermometry. Furthermore, significant entrainment of country rock in some samples played a major role during petrogenesis of the garnet- and cordierite-bearing granite but was not important during the evolution of the hornblende- and titanite-bearing granite. Intrusion of such hot, felsic magmas close to the inferred peak of metamorphism has probably caused, in part, the high temperature metamorphism and anatexis of the country rocks at relatively low pressures.

Global changes during Carboniferous–Permian glaciation of Gondwana: Linking polar and equatorial climate evolution by geochemical proxies

The most prevalent Phanerozoic glaciation occurred during the Carboniferous-Permian on the Southern Hemisphere Gondwana supercontinent. Sediments from the Pennsylvanian Dwyka Group deposited in the Karoo Basin of South Africa provide a complete record of glaciation and deglaciation phases. The direct correlation of glaciation events in southern Gondwana basins with the well-studied climate evolution of equatorial regions was previously hampered by lack of precise radiometric dating. As dating has now become available for the Karoo Basin, the Gondwana glaciation can be viewed in a global paleoclimatic framework with high temporal resolution. Element geochemical proxies (CIA [chemical index of alteration], Zr/Ti, Rb/K, V/Cr) record three confined shifts in climate and paleoenvironment of the Karoo Basin. These shifts were induced by changes in sea level, weathering rate, provenance, and redox conditions. Because of the low availability and diagenetic overprint of carbonates, ocean and atmosphere pCO 2 variations had to be reconstructed from δ 1 3 C o r g values of marine organic matter. The δ 1 3 C o r g signatures are affected by variable proportions of marine versus terrestrially derived organic matter and its state of preservation. Organic geochemical investigations (TOC [total organic carbon], C/N, lipid biomarkers) indicate the organic matter in the central Karoo Basin was primarily of algal origin. In agreement with element proxies, the varying δ 1 3 C o r g values mirror shifts in pCO 2 , rather than variations of organic-matter type. A covariation trend between carbon isotope signatures of equatorial carbonates and δ 1 3 C o r g values from the Karoo Basin argues against local forcing factors and instead implies a global climate-control mechanism. The 5-7 m.y. duration of a complete glacial cycle is not in tune with any known orbital frequency. Processes such as changes in equator-pole temperature gradients or newly developing atmosphere-ocean circulation pathways can be regarded as controlling factors.

Trace element and isotopic (Sr, Nd, Pb, O) arguments for a mid- to lower crustal origin of Pan-African garnet-bearing S-type granites from the Damara orogen (Namibia)

Geochronological data, major and trace element abundances, Nd and Sr isotope ratios, δ18O whole rock values and Pb isotope ratios from leached feldspars are presented for garnet-bearing granites (locality at Oetmoed and outcrop 10 km north of Omaruru) from the Damara Belt (Namibia). For the granites from outcrop 10 km N′ Omaruru, reversely discordant U–Pb monazite data give 207Pb/235U ages of 511±2 Ma and 517±2 Ma, similar to previously published estimates for the time of regional high grade metamorphism in the Central Zone. Based on textural and compositional variations, garnets from these granites are inferred to be refractory residues from partial melting in the deep crust. Because P–T estimates from these xenocrystic garnets are significantly higher (800°C/9–10 kbar) than regional estimates (700°C/5 kbar), the monazite ages are interpreted to date the peak of regional metamorphism in the source of the granites. Sm–Nd garnet–whole rock ages are between ∼500 and ∼490 Ma indicating the age of extraction of the granites from their deep crustal sources. For the granites from Oetmoed, both Sm–Nd and Pb–Pb ages obtained on igneous garnets range from ∼500 to ∼490 Ma. These ages are interpreted as emplacement ages and are significantly younger than the previously proposed age of ∼520 Ma for these granites based on Rb/Sr whole rock age determinations. Major and trace element compositions indicate that the granites are moderately to strongly peraluminous S-type granites. High initial 87Sr/86Sr ratios (>0.716), high δ18O values of >13.8‰, negative initial eNd values between −4 and −7 and evolved Pb isotope ratios indicate formation of the granites by anatexis of mid-crustal rocks similar to the exposed metapelites into which they intruded. The large range of Pb isotope ratios and the lack of correlation between Pb isotope ratios and Nd and Sr isotope ratios indicate heterogeneity of the involved crustal rocks. Evidence for the involvement of isotopically highly evolved lower crust is scarce and the influence of a depleted mantle component is unlikely. The crustal heating events that produced these granites might have been caused by crustal thickening and thrusting of crustal sheets enriched in heat-producing elements. Very limited fluxing of volatiles from underthrust low- to medium-grade metasedimentary rocks may have also been a factor in promoting partial melting. Furthermore, delamination of the lithospheric mantle and uprise of hot mantle could have caused localized high-T regions. The presence of coeval A-type granites at Oetmoed that have been derived at least in part from a mantle source supports this model.

Microfabrics indicating granulite-facies metamorphism in the low-pressure central Damara Orogen, Namibia

Abstract The reconstruction of the fluid evolution during progressive metamorphism in the central Damara orogen by Hoernes and Hoffer indicated that fluid-present conditions were only periodically attained among these rocks and that the highgrade metamorphism along the west coast of Namibia occurred under essentially water-absent conditions. The characteristic mineral assemblage in the metapelites is: plagioclase, quartz, alkali-feldspar, biotite, cordierite, garnet ± sillimanite, apatite, zircon, ore. Orthopyroxene-bearing parageneses in the metapelites are missing and granulite-facies metamorphism may not be detected on the basis of mineral assemblages. A detailed study along a 100 km traverse across the migmatite area in the central Damara Orogen (Namibia) reveals several microfabric features which indicate, in analogy to many other well known granulite-facies terrains where similar microfabrics have been described, that the Pan-African regional metamorphism in this area culminated at conditions of low-pressure granulite facies. The observed features are: 1. (a) The activation of basal 〈a〉- and prism 〈c〉-glide systems in quartz under conditions of plastic deformation, and the development of equivalent subgrain boundaries, parallel prism faces, and parallel basal planes. 2. (b) The orientated exsolution of rutile needles in quartz. 3. (c) The exsolution of string-perthites, mesoperthitic string-perthites and antiperthites in feldspars. 4. (d) In addition, the homogeneous distribution of major elements in garnet crystals and garnet-biotite thermometry suggest granulite-facies conditions.

Geochronology and petrology of migmatites from the Proterozoic Damara Belt - importance of episodic fluid-present disequilibrium melting and consequences for granite petrology

Abstract The Oetmoed Granite–Migmatite Complex (OGMC), Central Damara Orogen, Namibia, consists of grt- and crd-bearing S-type granites and hbl- and titanite-bearing A-type granites that intruded into crd-sil-Kfs-bearing metasedimentary rocks, stromatic migmatites and nebulites. Stromatic migmatites formed by limited in situ partial melting of metapelites under H2O-saturated conditions at ∼700°C and 4–5 kbar. This partial melting event took place close to the peak of regional metamorphism at ∼510 Ma as revealed by Sm–Nd garnet-whole rock ages as well as U/Pb monazite ages. The newly formed melt remained more or less at the site of origin. Melanosomes of the stromatic migmatites do not resemble true residues, instead they represent reaction zones between in situ melt and the metasedimentary host rock. Leucosomes from the stromatic migmatites are LREE- and HFSE-depleted which is typical for the low-melt fractions generally observed in migmatite terranes. Mass balance calculations suggest that these leucosomes may be disequilibrium melts. Similar δ 18 O whole rock values between 13‰ and 14‰ observed in leucosomes and corresponding melanosomes suggest homogenization of oxygen isotopes between the different rock types via the melt and an internal fluid phase during melting. Other leucosomes with a major element chemistry similar to the leucosomes from the stromatic migmatites but higher REE contents suggest significant incorporation of LREE-enriched phases (monazite) from the country rocks. For these leucosomes Sm–Nd garnet whole rock ages are ∼473 Ma indicating a second phase of melting. Nebulites mainly result from injection of granitic melts into the country rocks but their residual chemistry indicates that partial melting and limited melt removal must have occurred. Monazites from these nebulites record concordant U–Pb ages between ∼540 Ma and ∼470 Ma, indicating episodic migmatization during high-grade regional metamorphism. Intrusive peraluminous granites are likely generated by partial melting of pelitic sources in the lower crust, and different source rocks are probably not the controlling factor for the different chemistry of leucosomes and granites. Therefore, different melting conditions (fluid-absent vs. fluid-present) and different modes of entrainment and solubilities of accessory phases control the elemental budgets of the leucosomes. Large scale migmatite–granite complexes are a substantial part of some high-grade terrains but the distinctive geochemical composition of some leucosomes make these rocks unlikely precursors of large-scale granitic bodies.

Oxygen isotope thermometry based on a refined increment method and its application to granulite-grade rocks from Sri Lanka

Abstract In this paper we first present a revision of the increment method, a semi-empirical technique which allows the calculation of O-isotope fractionation factors between all silicates, provided their structure and chemical composition is known. For the temperature dependency, the experimentally calibrated high-pressure mineral-calcite fractionation factors of Chiba et al. (1989) are taken as a basis. Then this method is applied to granulite-facies parageneses from Sri Lanka including quartz, K-feldspar, sillimanite, garnet, pyroxenes, amphiboles, cordierite, magnetite and ilmenite. It can be shown that O-isotope equilibration occurred at granulite-facies conditions, at about 830°C, due to static recrystallization. Locally, retrograde re-equilibration occurred also at much lower temperatures due to late D4. In most cases, however, O-isotope disequilibria resulted from retrograde oxygen intracrystalline diffusion. It can be proved by model calculations that the O-diffusion data for water-saturated systems reported in the literature, are too high by two orders of magnitude to explain the measured isotopic compositions. The fractionation patterns allow the discrimination between open- and closed-system behaviour. The regional distribution of open- and closed-system patterns corresponds to the frequency of occurrence of late-stage migmatization (western part) and to differences in amount and composition of fluid inclusions.

Lower crustal melting and the role of open-system processes in the genesis of syn-orogenic quartz diorite-granite-leucogranite associations: constraints from Sr-Nd-O isotopes from the Bandombaai Complex, Namibia

Abstract The Bandombaai Complex (southern Kaoko Belt, Namibia) consists of three main intrusive rock types including metaluminous hornblende- and sphene-bearing quartz diorites, allanite-bearing granodiorites and granites, and peraluminous garnet- and muscovite-bearing leucogranites. Intrusion of the quartz diorites is constrained by a U–Pb zircon age of 540±3 Ma. Quartz diorites, granodiorites and granites display heterogeneous initial Nd- and O isotope compositions (eNd (540 Ma)=−6.3 to −19.8; δ18O=9.0–11.6‰) but rather low and uniform initial Sr isotope compositions (87Sr/86Srinitial=0.70794–0.70982). Two leucogranites and one aplite have higher initial 87Sr/86Sr ratios (0.70828–0.71559), but similar initial eNd (−11.9 to −15.8) and oxygen isotope values (10.5–12.9‰). The geochemical and isotopic characteristics of the Bandombaai Complex are distinct from other granitoids of the Kaoko Belt and the Central Zone of the Damara orogen. Our study suggests that the quartz diorites of the Bandombaai Complex are generated by melting of heterogeneous mafic lower crust. Based on a comparison with results from amphibolite-dehydration melting experiments, a lower crustal garnet- and amphibole-bearing metabasalt, probably enriched in K2O, is a likely source rock for the quartz diorites. The granodiorites/granites show low Rb/Sr ( 1) and are most likely generated by biotite-dehydration melting of heterogeneous felsic lower crust. All segments of the lower crust underwent partial melting during the Pan-African orogeny at a time (540 Ma) when the middle crust of the central Damara orogen also underwent high T, medium P regional metamorphism and melting. Geochemical and isotope data from the Bandombaai Complex suggest that the Pan-African orogeny in this part of the orogen was not a major crust-forming episode. Instead, even the most primitive rock types of the region, the quartz diorites, represent recycled lower crustal material.

Equilibrium relations of prograde metamorphic mineral assemblages

Abstractδ18O of quartz, biotite, muscovite, garnet, ilmenite or magnetite, K-feldspar, and δD of biotite from prograde metamorphic pelites of the Damara Orogen have been analyzed. The samples were taken from one stratigraphic unit which is exposed in the various stages from low-grade up to high-grade metamorphism with anatexis. Using the fractionation curves experimentally investigated up to now, it can be shown that equilibrium has been reached among the metamorphic assemblages in the low-grade and lower medium-grade metamorphism only. In the high-grade rocks only a partial equilibrium exists between those phases formed at the specific high-grade isogrades, i.e., K-feldspar, garnet, magnetite, or ilmenite, but not between these and the main reacting phase biotite. Biotite in equilibrium with these minerals is foundonly in anatectic rocks where new biotite crystallized from the melt.From this it is concluded that the18O/16O ratio of biotite is fixed at the time of crystallization and its initial composition is preserved when the temperature increases, unless the biotite recrystallizes. The isotopic temperatures derived from mineral pairs formed at specific isograds are in excellent agreement with temperatures derived from petrological investigations. The hydrogen isotoperatio of biotites decreases with increasing grade of metamorphism.

Petrology of basement-dominated terranes. II. Contrasting isotopic (Sr, Nd, Pb and O) signatures of basement-derived granites and constraints on the source region of granite (Damara orogen, Namibia)

Abstract Major and trace element and Nd, Sr, O and Pb isotope data from granites of the high-grade central part of the Damara orogen (Khan and Outlet gorge areas, Namibia) indicate a dominantly lower crustal origin. Based on their appearance in field, the granites can be divided into red granites, white granites and grey granodiorites. Red granites and some of the white granites from the Khan area are isotopically evolved (initial e Nd : −12.5 to −18.9) and were likely derived from metaigneous sources with late Archaean to early Proterozoic crustal residence ages. Other white granites are less evolved (initial e Nd : −6.5 to −8.8) and were likely derived from metasedimentary sources that are similar to the country rock metapelites. Grey granodiorites from the Khan and Outlet gorge area are also isotopically evolved (initial e Nd : −9.9 to −13.1) but are derived from metaigneous sources with younger, late Proterozoic crustal residence ages. Major and trace element data do not support closed-system fractional crystallization processes for all samples; however, some chemical features (i.e., decreasing Rare Earth Element (REE) abundances with increasing SiO 2 ) underline the importance of crystal fractionation processes for each distinct magmatic pulse. Isotope data do not support mixing of different crust-derived melts or assimilation of crustal rocks by a mafic magma. Instead, highly evolved Pb isotope compositions, strongly negative e Nd values and radiogenic Sr isotope compositions argue for an undepleted basement as a potential source. The most likely model involves mainly partial melting of different basement rocks of Archaean to Proterozoic age at different levels within the crust. Only some granites could be derived by melting of metasedimentary rocks of the Pan-African cover sequence. The consistency of the chemical data with a crustal anatectic origin and the range in radiometric ages suggests that they intruded simultaneously with crustal thickening; however, some of them may have been emplaced during extensional tectonics c. 40–50 Ma later than the main period of crustal thickening. The heating events that promoted melting of fertile deep-crustal rocks might have been caused either by the inferred high heat productivity together with crustal thickening during the main periods of orogeny, or by delamination of mantle lithosphere during the final extensional stages of the Pan-African orogeny.