How was the Bushveld complex assembled? Pulsed intrusions, crystallization conditions and the origin of layering in the Main Zone

点击翻译按钮获取中文摘要

The aim of this renewal project is to exploit new methods of automated spectroscopic drill core scanning in cooperation with the BGR Hannover. These can provide a much greater amount and quality of mineralogic information than anticipated in the ongoing project and will allow us to reach a more complete understanding of how magmas were added to the Main Zone magma chamber and how they evolved within it. The working hypothesis of the ongoing study was that the Main Zone was assembled from repeated magma injections from a deeper magma chamber, as evidenced in the northern lobe of the complex by cyclic vertical variations in physical properties (density variations, mineral abundances and compositions). Our studies of Main Zone drill cores in the eastern lobe, Marula site will test this hypothesis and provide a record of magma crystallization and differentiation by: (1) documenting rock density and mineral-chemical variations (2) measuring microtextural features of the main constituent minerals that reflect cooling rates and the extent of re-equilibration with residual melt; (3) calculating crystallization temperatures from trace-element and stable isotope geothermometry. The ongoing project succeeded in collecting density measurements on 272 samples regularly spaced along a 1500 meter section of the Main Zone. The density profile and preliminary mineral-chemical data suggest important differences compared with the northern lobe. Instead of regular, cyclic variations we found three contrasting subzones that guide sample selection for geochemical and microtextural studies in progress. The new studies in collaboration with the BGR in Hannover will involve scanning of the Main Zone samples using laser-induced breakdown spectroscopy (LIBS) and micro-XRF. If properly calibrated, these methods offer quantitative mineral proportions, 2D mineral distribution maps and multi-element concentration data. Importantly, the scanning is rapid and non-destructive, so all samples of the profile can be studied instead of a small subset originally planned. The spatial resolution of 20 microns for the microXRF will provide many analyses of the major mineral phases (plagioclase, ortho- and clinopyroxene) in a single sample. Statistical analysis will reveal the degree of chemical heterogeneity of each phase, which may be related via phase relations and crystallization models in basalt to changes in magma composition and/or temperature. The 2D element maps from microXRF will reveal mineral textures, zoning and the distribution and identity of late interstitial phases. We will also use of LIBS spectra to derive trace element concentrations in selected minerals that can track the chemical evolution of parental magmas.