ALBERT

Notes: Intrusion-related migmatites comprise a substantial part of the high-grade part of the southern Damara orogen, Namibia which is dominated by Al-rich metasedimentary rocks and various granites. Migmatites consist of melanosomes with biotite+sillimanite+garnet+cordierite+hercynite and leucosomes are garnet- and cordierite-bearing. Metamorphic grade throughout the area is in the upper amphibolite to lower granulite facies (5–6 kbar at 730–750 °C). Field evidence, petrographic observations, chemical data and mass balance calculations suggest that intrusion of granitic magmas and concomitant partial melting of metasedimentary units were the main processes for the generation of the migmatites. The intruding melts were significantly modified by magma mixing with in situ partial melts, accumulation of mainly feldspar and contamination with garnet from the wall rocks. However, it is suggested that these melts originally represented disequilibrium melts from a metasedimentary protolith. The occurrence of LILE-, HFSE- and LREE-enriched and -depleted residues within the leucosomes implies that both quartzo-feldspathic and pelitic rocks were subjected to partial melting. Isotope ratios of the leucosomes are rather constant (143Nd/144Nd (500 Ma): 0.511718–0.511754, ε Nd (500 Ma): −3.54 to −5.11) and Sr (87Sr/86Sr (500 Ma): 0.714119–0.714686), the metasedimentary units have rather constant Nd isotope ratios (143Nd/144Nd (500 Ma): 0.511622–0.511789, ε Nd (500 Ma): −3.70 to −6.93) but variable Sr isotope ratios Sr (87Sr/86Sr (500 Ma): 0.713527–0.722268). The most restitic melanosome MEL 4 has a Sr isotopic composition of 87Sr/86Sr (500 Ma): 0.729380. Oxygen isotopes do not mirror the proposed contamination process, due to the equally high δ18O contents of metasediments and crustal melts. However, the most LILE-depleted residue MEL 4 shows the lowest δ18O value (〈10). Mass balance calculations suggest high degrees of partial melting (20–40%). It is concluded that partial melting was promoted by heat transfer and release of a fluid phase from the intruding granites. High degrees of partial melting can be reached as long as the available H2O, derived from the crystallization of the intruding granites, is efficiently recycled within the rock volume. Due to the limited amounts of in situ melting, it seems likely that such regional migmatite terranes are not the sources for large intrusive granite bodies. The high geothermal gradient inferred from the metamorphic conditions was probably caused by exhumation of deep crustal rocks and contemporaneous intrusion of huge masses of granitoid magmas. The Davetsaub area represents an example of migmatites formed at moderate pressures and high temperatures, and illustrates some of the reactions that may modify leucosome compositions. The area provides constraints on melting processes operating in high-grade metasedimentary rocks.

Notes: The most widely used technique for the determination of high precision mineral growth ages in igneous and metamorphic rocks is dating of zircons with the U-Pb method. The interpretation of these ages, particularly in metamorphic settings, is hampered by an incomplete understanding of the common phenomenon of partial Pb-loss in zircon. In principle, this Pb-loss may occur in four very different ways: diffusion in metamict zircon, diffusion in pristine zircon, leaching from metamict zircon and recrystallization of metamict zircon. Here it is argued that, under conditions common in the continental crust, Pb-loss is only possible in partially to strongly metamict zircons. Pb-diffusion in the pristine zircon lattice is insignificant up to temperatures of at least 1000 °C. Pb-loss is only possible if the zircons experienced a time interval below their annealing temperature of about 600–650 °C, because only below this temperature can the lattice damage through α-decay and spontaneous fission accumulate. Zircons that remain above this temperature do not lose Pb by diffusion and will stay closed systems. Complete resetting of the U-Pb system in zircon under crustal conditions is only possible through dissolution and reprecipitation of zircon. Partial resetting results from recrystallization, leaching or diffusion in metamict zircon. As a consequence, special care has to be taken to interpret lower intercepts on concordia diagrams defined by discordant U-Pb data. Lower intercept ages may be significant only if they are defined by zircons with low U-content (〈100 p.p.m.) or if confirmed by other geochronological methods. In addition, the accuracy of the lower intercept should be confirmed by abrading the zircon fractions that define the discordia.