Journal of Petrology, Volume 39, Issue 02: February 1 1998.
The Goboboseb-Messum volcanic centre is the source of two voluminous silicic eruptive sequences, the Goboboseb quartz latites (Units I-III), and the Springbok quartz latite unit (both within the Awahab Formation). Intrusive equivalents exist as plugs and a laccolith peripheral to the Messum Complex. The recognition of correlatives of these quartz latite units in the southeastern Paraná suggests eruptive volumes of ~2320 km3 (Goboboseb units) and 6340 km3 (Springbok unit). The latter is thought to be a single eruptive event. Phenocryst assemblages are plagioclase (An51-63), pyroxene, titanomagnetite and apatite. Pyroxene assemblages range from augite, to augite + pigeonite, to pigeonite, to pigeonite ± hypersthene, the assemblages changing progressively from the Goboboseb unit through to the Springbok unit. Although pyroxene phenocrysts from individual samples are compositionally very uniform there is a small increase in Fe through the sequence, attributed to decreasing temperature (± pressure). Thermometry suggests melt temperatures >1000°C. Many plagioclases contain abundant glass inclusions of three compositional types, thought to result from active disequilibrium melting at magma chamber walls. Relatively small, but systematic, changes in whole-rock composition occur stratigraphically from the lowest Goboboseb unit through to the Springbok unit and to their correlatives in the Paraná, best shown by SiO2 (67-71%) and FeO* (5·4-7·4%) which increase and decrease, respectively, in the progressively younger eruptive phases. P, Ti, Y, Zr, Nb and Cu are positively correlated with FeO*, whereas [epsilon]Sr and Pb isotope compositions correlate inversely with FeO*. Crust-normalized spidergram plots indicate strong negative Sr anomalies, accompanied by significant Eu/Eu* anomalies (0·62-0·67). The quartz latite melts can be interpreted in terms of large-scale assimilation-fractional crystallization (AFC)-style processes, involving high degrees of lower- and upper-crustal melting, with thermal and material input from hybridized LTZ.L-type basaltic magmas (Part I). Thermal source is inferred to be the Tristan plume. The crustal end-member is thought to be the mid-Proterozoic restite source of the Damara granites, although some shallower crustal input is also likely. Modelling suggests the source may be similar to A-type granites and charnockites (i.e. relatively REE and HFSE enriched). Available seismic data suggest a simple velocity crustal profile, possibly the result of the massive crustal and uppermost mantle melting that accompanied the evolution of the Awahab magma system.
Keywords:
Pages 227-253