Greenhills revisited : an investigation into the geochemistry of the Greenhills Ultramafic Complex

Webster, Glen Robert.

The Greenhills Ultramafic Complex lies about 12km to the south-west of Invercargill and occupies approximately 15km2 of the n~rthernmost corner of the Bluff Peninsula. It is part of the Brook St terrane" and is believed to represent the roots of a Late Permian island arc volcanic chain that has had its upper portion subsequently removed by erosion. Mossman(1970) undertook a comprehensive study of the area as the basis of a PhD thesis and concluded that the Greenhills Ultramafic Complex represents the remains of a shallow level magma chamber that was open to the atmosphere and of a hydrous nature. The Greenhills Ultramafic Complex consists of a layered structure consisting of ultramafic rocks at the base and a gabbroic portion at the top. The sequence of minerals observed is consistent with fractionation of a basaltic magma with olivine predominating and being accompanied by residual chromite in the first rock type to occur: dunite. Later in the sequence clinopyroxene begins to crystallise and the rock type becomes a wehrlite. Next in the sequence of minerals is orthopyroxene, however its modal proportion is far lower than most other layered ultramafic complexes. Skaergaard (Wager and Deer 1939)contained websterite rather than wehrlite the difference being the predominance of orthopyroxenes in the rock rather than the clinopyroxenes as occurs at Greenhills. Repeated intrusions of more felsic magma as dikes up through the base and into the intrusion pre-empted the fractional crystallisation that was occurring so that an upper gabbroic layer was deposited at the top of the sequence. Mossman concluded that the composition of the magma was basaltic and that it had affinities between calc-alkaline and tholeiitic, which at the time of his investigation was quite an innovative idea. ~] _]! _] -] -] J J J J J J J ] J l l This suggestion is investigated to try to determine whether this is a viable source magma for the observed fractionation sequence and the conclusion is that it is a viable mechanism for forming the minerals that comprise the complex. The difference between this study and that of Mossman is that it is suggested here that the corrugated contact between the wehrlite and dunite was caused by small scale convection cells eroding the base of the magma chamber and depositing the debris by gravitational settling elsewhere in the magma chamber and this process results in the rhythmic layering observed here . This is a similar, process to that which occurred in the Skaergaard intrusion (Irvine 1987) except that the small convection cells do not occur as a consequence of double diffusive convection but simply that the magma chamber has too low a roof to allow the development of a convection cell that is the full width of the magma chamber. A time constraint has been put on the age of the intrusion by some recent work and the collection of ages for surrounding metasediments allows a clearer picture top be obtained on how and when the complex developed. The geochemical bias of this project has meant that confirmation could be obtained for multiple intrusions of magma throughout the complex because in the past the evidence had been entirely petrographical. A clearer understanding of how the Greenhills Ultramafic Complex developed and of the processes that occurred inside it while it was still crystallising.

98 leaves, [1] folded leaf of plates : ill. (some col.), col. map ; 30 cm.

1994Webster