Abstract
The southernmost limit that the Hikurangi Plateau subducted beneath New Zealand in the Early Cretaceous has been interpreted from seismic data to be somewhere between Dunedin and Timaru. I test this hypothesis from a petrological perspective using garnet pyroxenite xenoliths from the Kakanui Mineral Breccia in North Otago. These xenoliths represent some of the only mafic material that has been entrained and brought to the surface from depth beneath New Zealand and therefore may be part of the Hikurangi Plateau. Geochemical analysis has been applied to 6 samples. Calculated equilibration conditions range from between 1081 and 1260oC at ~45 km depth, which suggest equilibration was attained within the mantle lithosphere. Whole rock major element, trace element and REE data show the garnet pyroxenite xenoliths to have compositions that broadly correspond to MORB-like. This is similar, but not identical, to the main (plateau) portion of the Hikurangi Plateau. However, radiogenic isotope data obtained from garnet and clinopyroxene separates show the xenoliths have MORB-like chemistries that are distinct from the OIB-like isotopes of the main basaltic portions of the Hikurangi Plateau. The garnet pyroxenite Pb isotope data are not radiogenic enough for the xenoliths to represent intraplate basalt melts or cumulates that stalled and were metamorphosed in the mantle. The garnet pyroxenites could represent metamorphosed MORB from the base of the plateau, but unpublished Lu-Hf isotope data indicates that they are much older and were metamorphosed in the Cambrian (~580-590 Ma). Therefore, these xenoliths represent fragments of MORB-like melts long stored in the lithospheric mantle underneath Kakanui. There remains no petrological evidence for the Hikurangi Plateau under the central southern South Island.