The 'Ins' and 'Outs' of the Bushveld Complex Upper Zone

This dissertation is an investigation into the geochemical and geodynamic evolution of the Upper Zone of the Bushveld Complex. The Bushveld Complex is one of the few large layered intrusions in which the entire cumulate stratigraphy is preserved and well-exposed from its base to its roof. Despite this unique feature, relatively little is known about the nature of magmatic differentiation in the uppermost portions of the Bushveld. As a first order, I quantify the chemical composition of the preserved stratigraphy (i.e. the bulk composition) from the geochemical base of the Upper Zone (the Pyroxenite Marker) to the contact with the roof. On the basis of major element modeling and trace element equilibria I show that the bulk composition is not representative of the original magma composition, and therefore some magma must be missing. I propose that the Rooiberg Group lavas and/or Rashoop granophyres, which make up the immediate roof of the intrusion, represent the missing magma.;A further test of the magma-loss hypothesis comes from the trace element contents of apatites in the uppermost 625 m of the Upper Zone stratigraphy. Comparison of the equilibrium liquid compositions calculated from these apatites with the Rooiberg and/or Rashoop roof rocks shows that they are a geochemical match. My results also indicate a role for large-scale (>625 m) liquid immiscibility at the top of the Bushveld. This is the first documented evidence for liquid immiscibility based on the compositions of mineral phases, not melt inclusions; and it is the first quantitative evidence for large-scale immiscibility in the Bushveld Complex.;Quantification of the parent magma composition at the Pyroxenite Marker allows me to not only estimate the 'outputs' from the magma chamber, but also to constrain the 'inputs'. A geochemical record of magma input and mixing is recorded in the cumulate stratigraphy for approximately 350 m below the Pyroxenite Marker. Using the evolution in mineral compositions I calculate the composition and proportion of incoming magma to the Upper Zone, as well as the style of input. The composition of the incoming magma is then compared to other known pulses of magma into the Bushveld Complex in order to put constraints on the source contributions and formation dynamics of the intrusion as a whole.