|dc.description.abstract||The South Auckland volcanic field (SAVF) is situated in a continental intraplate tectonic setting and consists of silica-undersaturated alkalic basalts that erupted from about 100 centres between 1.59 and 0.51 Ma. The basalts range in composition from basanites to quartz tholeiites. Two basalt groups are distinguished based on their contrasting mineralogical and geochemical characteristics: group A, a less silica-undersaturated transitional basalt to quartz tholeiite suite; and group B, a strongly silica-undersaturated basanite to nepheline hawaiite suite. The petrogenesis and evolution of the SAVF basalts is examined using new data from 200 samples on mineral chemistry from electron microprobe analyses, major and trace elements from XRF, incompatible trace elements and REE from ICP-MS, and Sr, Nd, and Pb isotopic data from TIMS analyses.
Forsteritic olivine is the dominant phenocryst phase in each rock group (group A: Fo₆₁ - Fo₈₂; group B: Fo₆₄ - Fo₉₂). Clinopyroxene is abundant and ranges from augite (Wo₃₅₋₄₇ En₃₆₋₅₀ Fs₁₁₋₂₃) in group A to diopside (Wo₄₂₋₅₂ En₃₂₋₄₆ Fs₁₀₋₂₀) in group B basalts. Plagioclase (An₄₀₋₇₀) is found as a phenocryst only in group A basalts. The group A basalts are distinguished by their low total alkalis (3.0 - 4.8 wt.%), Nb (9 - 29 ppm) and Zr (97 - 210 ppm) abundances, and low (La/Yb)N (3.4- 7.6), which contrast with group B that have large total alkalis (3.3 - 7.9 wt.%), Nb (35 - 102 ppm) and Zr (194 - 491 ppm), and (La/Yb)N (12 - 47).
The incompatible trace elements of group A lavas have characteristics indicating derivation from an enriched upper mantle source but relatively depleted in Th, K, Nb, Ta, and LREE with small LREE/HREE values, whereas those in group B exhibit strong trace element affinities (i.e., large Th, Nb, Ta, and LREE abundances) with an OIB-like source. Petrogenetic modelling suggests that the group A and B lavas evolved as discrete lineages that do not appear to be related to a common parental magma or source. Partial melting models indicate that < 8 % melting of a garnet-peridotite source is required to produce primary group B magmas, whereas primary group A magmas are generated from < 11 % melting of a spinel-peridotite source. Differentiation of basalts from each group is dominated by olivine and clinopyroxene fractionation, and only minor plagioclase in group A.
Both rock groups fall within a narrow range of Sr and Nd isotopic compositions (⁸⁷Sr/⁸⁶Sr= 0.70273±19 - 0.70330±17 and εNd = + 5.97 to + 6.89) similar to the composition of HIMU-OIB, whereas Pb isotopic compositions are unradiogenic relative to HIMU-OIB (i.e., ²⁰⁶Pb/²⁰⁴Pb = 18.95 - 19.33; ²⁰⁷Pb/²⁰⁴Pb = 15.586 - 15.589; ²⁰⁸Pb/²⁰⁴Pb = 38.731 -38.906) and range between Atlantic MORB and EMII. The group A lavas have slightly higher ⁸⁷Sr/⁸⁶Sr and ²⁰⁷Pb/²⁰⁴Pb, lower ²⁰⁶Pb/²⁰⁴Pb and ²⁰⁸Pb/²⁰⁴Pb, and similar εNd values compared to those in group B, but there is no isotopic or geochemical evidence to suggest that any of the lavas have been contaminated by continental crust. Therefore, the relatively small but distinct differences in Sr and Pb isotopic compositions and similar εNd values are considered to be source-related.
Variations in incompatible element ratios such as K/Nb and Zr/Nb indicate that the SA VF basalts were derived from two distinct sources. Group B basalts show incompatible trace element signatures characteristic of a LREE-enriched HIMU-OIB-like source, which can be accounted for by recycling of subducted oceanic crust at mantle depths. In contrast, the group A basalts have incompatible trace element affinities that are intermediate between a HIMU and an EMII-type component that could have resulted from subduction-related metasomatism of the subcontinental lithosphere modified by a HIMU plume. These events may have been connected with the subduction of the Phoenix plate and plume-related magmatism when New Zealand was at the eastern margin of the Gondwanaland supercontinent.||