Margins resulting from continental breakup are generally classified as volcanic (related to flood basalt volcanism from a starting plume head) or non-volcanic (caused by tectonic processes), but many margins (breakups) may actually be hybrids caused by a combination of volcanic and tectonic processes. It has been postulated that the collision of the Hikurangi Plateau with the Gondwana margin similar to 110 Ma ago caused subduction to cease, followed by large-scale extension and ultimately breakoff of the Zealandia micro-continent from West Antarctica through seafloor spreading which started at similar to 85 Ma. Here we report new geochemical (major and trace element and Sr-Nd-Pb-Hf isotope) data for Late Cretaceous (99-69 Ma) volcanism from Zealandia, which include the calc-alkalic, subduction-related Mount Somers (99-96 Ma) and four intraplate igneous provinces: 1) Hikurangi Seamount Province (99-88 Ma), 2) Marlborough Igneous Province (98-94 Ma), 3) Westland Igneous Province (92-69 Ma), and 4) Eastern Chatham Igneous Province (86-79 Ma). Each of the intraplate provinces forms mixing arrays on incompatible-element and isotope ratio plots between HIMU (requiring long-term high mu = U-238/Pb-204) and either a depleted (MORB-source) upper mantle (DM) component or enriched continental (EM) type component (located in the crust and/or upper mantle) or a mixture of both. St. Helena end member HIMU could be the common component in all four provinces. Considering the uniformity in composition of the HIMU end member despite the type of lithosphere (continental, oceanic, oceanic plateau) beneath the igneous provinces, we attribute this component to a sublithospheric source, located beneath all volcanic provinces, and thus most likely a mantle plume. We propose that the plume material rose beneath the active Gondwana margin and flowed along the subducting lithosphere beneath the Hikurangi Plateau and neighboring seafloor and through slab tears/windows beneath the Gondwana (later to become Zealandia) continental lithosphere. We conclude that both plateau collision, resulting in subduction cessation, and the opening of slab tears/windows, allowing hot asthenosphere and/or plume material to upwell to shallow depths, were important in causing the breakup of Zealandia from West Antarctica. Combined tectonic-volcanic processes are also likely to be responsible for causing breakup and the formation of other hybrid type margins. (C) 2019 Elsevier B.V. All rights reserved.

Data from New Zealand and northern Victoria Land, Antarctica, indicate that the Cambrian Takaka Terrane intra-oceanic arc/backarc assemblage and the Bowers Terrane intra-oceanic arc/back-arc assemblage were accreted to the Gondwana margin by the Late Cambrian. Compelling similarities between the arc rocks and the immediate post-arc sediments firmly place the two regions in the same tectonic framework and imply close paleogeographic proximity. Currently, the Ross Orogen is thought to be the result of sinistral oblique convergence with west-directed subduction, and accretion of the arc assemblages is attributed to closure of backarc basins. Syntectonic fluvial conglomerates in both regions attest to the development of fluvial systems draining both the accreted arc and the contemporaneous continental margin arc. Trilobite faunas indicate that fluvial sedimentation commenced earlier in New Zealand than in northern Victoria Land. In the context of the widely accepted sinistral oblique convergence model, these data suggest an original position for New Zealand to the south of northern Victoria Land, probably in the region of the southern Ross Sea.

Coupled O and Hf isotopic compositions of zircons from Early Cretaceous (113-124 Ma) granitoids of the Separation Point Suite (SPS), New Zealand, obtained by cathodoluminescence imaging-guided micro-beam methods (SIMS, LA-ICPMS), are used as a record of evolving magma compositions in a prominent Mesozoic arc system. Eight representative SPS samples from individual plutons in the Nelson and Fiordland regions yield magmatic zircons with initial Hf isotope ratios (expressed in epsilon(Hf)) ranging from -4 to + 11 (Nelson) and + 5 to + 12 (Fiordland), respectively. Initial Hf isotope ratios of zircons are extremely heterogeneous within individual samples, with the vast majority of values distinctly less radiogenic than depleted mantle at similar to 120 Ma (epsilon(Hf)similar to+ 16). O isotope ratios are likewise variable, with delta O-18(zircon)(SMOW) values of 2-8%omicron (Nelson) and 0-7%omicron (Fiordland). The within-rock variability in both zircon Hf and 0 isotope ratios is testimony to open-system processes that operated during magma evolution and zircon crystallisation. Average delta O-18(zircon) for cores and rims allow constraints to be placed on the O isotopic composition of magmas from which zircon precipitated (delta O-18(magma)similar to 4-8%omicron). Elevated delta O-18(magma) (>6.5%omicron) require involvement of O-18-enriched supracrustal material (weathered continental crust or low-T seawater-hydrothermally altered oceanic crust), while delta O-18(magma)<5.5%omicron imply contribution from a O-18-depleted crustal component. Whole rock Sr isotope, Nb/Ta and NdfPb systematics are inconsistent with 180-depleted slab melts (618O similar to 0-6%omicron)as a source component for SPS magmas. Instead, low delta O-18(magma) values suggest incorporation of high-T meteoric-hydrothermally altered country rocks similar to those of the Largs terrane presently exposed in northern Fiordland. In diagrams Of SiO2 versus inferred delta O-18(magma) the most primitive SPS samples from the Nelson and Fiordland regions plot close to the expected composition for primitive arc magmas. More evolved granitoids, however, show strongly divergent trends of O isotope composition as a function Of SiO2, suggestive of assimilation coupled to fractional crystallisation (AFC) in contrasting crustal environments. Emplacement-level contamination by local crust is supported by age distributions of inherited zircons, which indicate a predominance of Palaeozoic and Mesozoic zircons in Nelson and Fiordland granitoids, respectively. (c) 2008 Elsevier B.V. All rights reserved.

Pyroclastic fall deposits of the paired Rotoiti and Earthquake Flat eruptions from the Taupo Volcanic Zone (New Zealand) combine to form a widespread isochronous horizon over much of northern New Zealand and the southwest Pacific. This horizon is important for correlating climatic and environmental changes during the Last Glacial period, but has been the subject of numerous disparate age estimates between 35.1±2.8 and 71±6ka (all errors are 1s.d.), obtained by a variety of techniques. A potassium–argon (K–Ar) age of 64±4ka was previously determined on bracketing lavas at Mayor Island volcano, offshore from the Taupo Volcanic Zone. We present a new, more-precise 40Ar/39Ar age determination on a lava flow on Mayor Island, that shortly post-dates the Rotoiti/Earthquake Flat fall deposits, of 58.5±1.1ka. This value, coupled with existing ages from underlying lavas, yield a new estimate for the age of the combined eruptions of 61.0±1.4ka, which is consistent with U–Th disequilibrium model-age data for zircons from the Rotoiti deposits. Direct 40Ar/39Ar age determinations of plagioclase and biotite from the Rotoiti and Earthquake Flat eruption products yield variable values between 49.6±2.8 and 125.3±10.0ka, with the scatter attributed to low radiogenic Ar yields, and/or alteration, and/or inheritance of xenocrystic material with inherited Ar. Rotoiti/Earthquake Flat fall deposits occur in New Zealand in association with palynological indicators of mild climate, attributed to Marine Isotope Stage (MIS) 3 and thus used to suggest an age that is post-59ka. The natures of the criteria used to define the MIS 4/3 boundary in the Northern and Southern hemispheres, however, imply that the new 61ka age for the Rotoiti/Earthquake Flat eruption deposits will provide the inverse, namely, a more accurate isochronous marker for correlating diverse changes across the MIS 4/3 boundary in the southwest Pacific.

New SHRIMP U-Pb ages and geochemical data have been obtained for the volcano-sedimentary Loch Burn Formation (LBF). A rhyolitic clast from the tops of the Stuart Mountains gave a SHRIMP age of 150.3 +/- 11.9 Ma, and a very fine sandstone from the same area was dominated by 147.9 +/- 2.1 Ma zircons. These ages imply a <148 Ma depositional age for the LBF in this area, in contrast to a previous 195(-1)(+3) Ma age determination for an intercalated felsic flow in the North Fiord area, c. I km distant. Two tonalitic clasts from Cumbrae Island gave ages of 354.6 +/- 2.6 and 326.8 +/- 3.2 Ma, respectively. The c. 47 m.y. difference between depositional ages in the Stuart Mountains and North Fiord indicates that the LBF as currently mapped includes more than one unit. However, volcanic clasts and lavas from the two areas are indistinguishable in terms of major and trace element geochemistry, and there is currently insufficient information to discriminate between the two units. Informal units are therefore proposed: the <148 Ma LBF-2 unit and the c. 195 Ma LBF-1 unit. Volcanic clasts and lavas from both units are probably derived from the Darran Suite arc. The c. 327 Ma tonalite clast has moderate Zr/TiO2 and chemistry similar to most of the other LBF tonalite samples, whereas the c. 355 Ma tonalitic clast has unusual, very high Zr chemistry. Neither appear to be related to any New Zealand plutonics currently well characterised in the literature, based on differences in major and trace element chemistry. However, recent work indicates that possible correlatives with appropriate age and geochemistry exist for both groups of tonalites. No Western Province material is recorded in any of the LBF samples, but this is not considered to exclude formation adjacent to or within the Gondwana margin.

In order to understand trace element behaviour during combustion of coals from the Greymouth coalfield, combustion tests were performed on three seam composite samples. The major and trace elements from sub-samples of feed coal, bottom ash, fly ash, and flue gas were analysed by different techniques including inductively coupled plasma mass spectrometry (ICP-MS), inductively coupled plasma-atomic emission spectrometry (ICP-AES), wavelength dispersive X-ray fluorescence (WD-XRF), and scanning electron microscopy with energy-dispersive X-ray analyser (SEM-FDXA). To help better understand trace element partitioning in combustion ash, float-sink and sequential leaching experiments were also employed to determine the association of trace elements with mineral matter or organic matter. Instrumental Neutron Activation Analysis (INAA) was also employed to determine trace element content in float and sink fractions of fly ash as well as in three major phases in the bottom ash.The partitioning behaviour of trace elements, including some that are environmentally sensitive, was also investigated through the use of float-sink tests and direct determination of trace elements in different combustion ash types and phases, Mass balance and partitioning of major and trace elements have been studied to determine the fate of trace elements after combustion, The partitioning of trace elements, especially hazardous air pollutants (HAPs). in different combustion ashes. can be summarised as follows:1. Most trace elements, especially As, Ba, Co, Cr, Mn, Ni, are partitioned in the glassy and refractory bottom ash fractions.2. A significant proportion of trace elements of As, Se, and Pb are partitioned into fly ash fractions.3. Some volatile elements (e.g. > 90% of S and Hg and up to 64% of Cl) and, to a lesser extent. B (up to 44%) and Cd (up to 50%) are partitioned in the flue gas fraction.4. Although the low ash yield of Greymouth coal seams have the advantage of generating less solid combustion ash, one of the accompanying consequences is that resultant trace elements tend to be enriched in the ash to a greater magnitude than other more clastic sediment influenced coals. (c) 2005 Elsevier B.V. All rights reserved.

The Late Cretaceous-Paleocene succession exposed on the Tora coast, near the southeastern tip of the North Island, is distinguished by an unusual lithofacies of the Whangai Formation, and by an apparently unique formation, Manurewa Formation, which spans the Cretaceous/Tertiary (K/T) boundary.The Late Cretaceous siliceous Whangai Formation at Tora includes zones of slumps and olistostromes, containing megaclasts of limestone up to 3 m long. The olistostromal deposits suggest steep submarine topography with a high rate of erosion, and imply tectonic activity. The common occurrence of hummocky cross-stratification suggests deposition in shelf depths above storm wave base. The sharply overlying Manurewa Formation is interpreted as the infill of a major shallow channel complex, perhaps >9 km wide and spanning the K/T boundary in time. The older of two channelled units is of latest Cretaceous (latest Haumurian/late Maastrichtian) age, and consists of bioturbated alternating thin sandstone and mudstone with thin conglomerate lenses and limestone beds. It is likely to have been deposited in a low-energy environment, probably deeper than that of the Whangai. The younger channel system, of early Paleocene (early Teurian) age, erodes into the older in the northeast, and into the underlying Whangai Formation in the southwest. Basal deposits consist predominantly of medium to coarse, thick-bedded, glauconitic sandstone, with local low-angle cross-stratification and microflora typical of low salinity conditions, suggesting deposition in shallow shelf depths. These deposits contain olistrostromes with megaclasts up to 1 m long of limestone and rarer dark grey siltstone or very fine sandstone clasts typical of Whangai Formation. The inclusion of megaclasts of Whangai Formation indicates that local emergence and erosion of older strata was occurring. Deposits grade upward into well-sorted bioturbated sandstones of the Awhea Formation, with prominent low-angle cross-stratification, interpreted as very shallow marine, probably nearshore deposits.The channel system represented by the Manurewa Formation records an initial relative sea-level rise, followed by an abrupt sea-level fall at, or close to, the K/T boundary. New Zealand was in a passive margin tectonic setting at the time, but the widespread presence of olistostromes, some including clasts derived from older strata, suggest that local tectonic activity and uplift was occurring. The effects may have been enhanced by a climatic shift in storm tracks and intensity in the latest Cretaceous, which is supported by the evidence of strong wave activity.By contrast, to the south in Marlborough, the K/T boundary succession is commonly characterised by an apparently conformable lithologic change from limestone to chert, although with local hiatus. To the north, in southern Hawke's Bay, the coeval succession is characterised by a disconformity separating greensand from underlying light grey, slightly calcareous mudstone of the Whangai Formation. The Tora sequence may provide the link between two distinctly different lithologic successions.

Leaching processes are believed to be responsible for the unusually low-ash content (sometimes less than 1%) of the thick (up to 35 m) Cretaceous coals located in the Greymouth coalfield, South Island, New Zealand. Although leaching of inorganics in peat is a generally accepted process, little is known about leaching after burial. The “Main” and “E” seams in the Greymouth coalfield show good correlation between low ash and bed thickness. The ash content, however, is often less than 1%, which is lower than most known modern analogues (i.e. peat). There are several lines of evidence that suggest that mineral matter may have been removed from the coal not only in the peat stage but also after burial. For example, etching features found in quartz grains and clay aggregates indicate that some leaching processes have taken place. In addition, liptinitic material (e.g., bitumen) in the cleat networks supports the conclusion that there has been some movement of solutions through the coal after burial. These solutions may have helped to remove some of the inorganics originally within the Greymouth coals.

What is believed to be a very unusual mode of occurrence for lead in coal has been identified as crocoite (PbCrO4). As part of a larger study on trace elements and mineralogy in the Cretaceous Main Seam in New Zealand, crocoite was found in raw coal samples within the lower part of the coal seam. X-ray diffraction (XRD) and bulk chemical data from a SEM equipped with an energy dispersive X-ray analyser (EDXA) have confirmed the identity of this mineral. This is apparently the first time that crocoite has been reported in coal. Crocoite usually occurs only in the oxidised zone of lead mineral deposits. The occurrence of this mineral in the Main Seam coal implies that the deposit was exposed to an oxidising environment at some stage, most likely after coalification.

The 0.23 Ma Kaingaroa Ignimbrite is a composite, multiple flow-unit ignimbrite erupted from the Reporoa Caldera, Taupo Volcanic Zone, New Zealand. Reporoa Caldera evolved from a single vent during the initial stages of the Kaingaroa eruption, to a trapdoor caldera with the opening up of fractures along the eastern margin and asymmetric eruption of early phases of the ignimbrite. Finally, plate collapse occurred during the later stages of the eruption. Kaingaroa Ignimbrite pumice clasts range in composition from dacite to rhyolite. Five pumice types have been identified based largely on the geochemical variation of Rb and Sr and ferromagnesian mineral composition. Pumice types A, B, C, and D exhibit variations in mineralogy, trace element chemistry, and isotopic composition consistent with derivation from a weakly zoned magma chamber. A subordinate dacitic juvenile component (type E) appears to be unrelated to the Kaingaroa pumices by crystal fractionation or assimilation, and is interpreted as a distinct magma batch that was incorporated into the eruption. The asymmetrical removal of magma from the chamber during caldera collapse is inferred to have resulted in a greater degree of drawdown along the eastern margin of the chamber, as suggested by the current spatial distribution of post-caldera rhyolites.

New helium isotope data measured in Cenozoic intraplate basalts and their mantle xenoliths are compared with present-day mantle helium emission on a regional scale from thermal and nonthermal gas discharges on the South Island of New Zealand and the offshore Chatham Islands. Cenozoic intraplate basaltic volcanism in southern New Zealand has ocean island basalt affinities but is restricted to continental areas and absent from adjacent Pacific oceanic crust. Its distribution is diffuse and widespread, it is of intermittent timing and characterised by low magma volumes. Most of the He-3/He-4 ratios measured in fluid inclusions in mantle xenocrysts and basalt phenocrysts such as olivine, garnet, and amphibole fall within the narrow range of 8.5 +/- 1.5 Ra (Ra is the atmospheric He-3/He-4 ratio) with a maximum value of 11.5 Ra. This range is characteristic of the relatively homogeneous and degassed upper MORE-mantle helium reservoir. No helium isotope ratios typical of the lower less degassed mantle (>12 Ra), such as exemplified by the modem hot-spot region of Hawaii (with up to 32 Ra) were measured. Helium isotope ratios of less than 8 Ra are interpreted in terms of dilution of upper mantle helium with a radiogenic component, due to either age of crystallisation or small-scale mantle heterogeneities caused by mixing of crustal material into the upper mantle. The crude correlation between age of samples and helium isotopes with generally lower R/Ra values in mantle xenoliths compared with host rock phenocrysts and the in general depleted Nd and Sr isotope ratios and the light rare earth element enrichment of the basalts supports derivation of melts as small melt fractions from a depleted upper mantle, with posteruptive ingrowth of radiogenic helium as a function of lithospheric age.In comparison, the regional helium isotope survey of thermal and nonthermal gas discharges of the South Island of New Zealand shows that mantle He-3 anomalies in general do not show an obvious relationship with either age or proximity to the Cenozoic intraplate volcanic centres or with major faults. In general, areas characterised by mantle He-3 emission are interpreted to define those regions beneath which mantle melting and basalt magma addition to the crust are recent. The strongest mantle He-3 anomaly (equivalent to >80% mantle helium component) is centred over southern Dunedin, measured in magmatic CO2-rich mineral water springs issuing from crystalline basement rocks which outcrop at the southern extent of Miocene intraplate basaltic volcanism which ceased 9 Ma ago. This mantle helium anomaly overlaps with an area characterised by elevated surface high heat flow, compatible with a long-lived mantle melt/heat input into the crust. In comparison Banks Peninsula, another Miocene intraplate basaltic centre, is characterised by relatively low surface heat flow and a small mantle helium contribution measured in a nitrogen-rich spring. Here the thermal transient induced by the magmatic event has either dissipated or has not reached the surface. In the former case one might be dealing with storage and mixing of magmatic and crustal gases at shallow crustal levels and in the latter with active to recent mantle-melt degassing at depth. Along the most actively deforming part of the plate boundary zone, the transpressional Alpine Fault and Marlborough fault systems, mantle helium is present in gas-rich springs in all those areas underlain by actively subducting oceanic crust (the Australian plate in the south and Pacific plate in the north), whereas the central part of the Alpine transpressional fault is characterised by pure crustal radiogenic helium. Areas where the mantle helium component is negligible are restricted to the centre part of the South Island, extending along its length from Southland to northern Canterbury and Murchison. These areas are interpreted to delineate the extent of thicker and colder lithosphere compared to all other areas where mantle helium release from partial mantle melts at depth is recent to active being added to the lower lithosphere and/or lower crust. Areas characterised by mantle helium anomalies are equated with areas of thermal mantle anomalies, i.e., localised mantle heterogeneities such as upwelling less dense silicate melts in the upper asthenospheric mantle. Copyright (C) 2000 Elsevier Science Ltd.

The 0.89 Ma Tikorangi Ignimbrite (revised name) is a mixed andesite-rhyolite ignimbrite preserved within a localised area of the Matahana Basin on the western side of the Taupo Volcanic Zone, New Zealand. The source of the ignimbrite is poorly constrained, but the location and limited maximum pumice and lithic data available suggest that the ignimbrite was emplaced by a pyroclastic flow erupted from within the Kapenga caldera complex. The ignimbrite is the oldest exposed unit sourced from the complex and offers a window into the early history of the Taupo Volcanic Zone.The Tikorangi Ignimbrite can be conveniently divided into three units. The Lower Tikorangi ignimbrite (LTi) is dominantly rhyolitic and has been subject to varying degrees of hydrothermal alteration This progressively grades into the Middle Tikorangi ignimbrite (MTi), which best illustrates the mixed nature of the ignimbrite, and has four types of juvenile pumice: black, grey black, brown black (all andesitic, in various states of oxidation), and white-grey (rhyolitic). The top of the Middle Tikorangi ignimbrite grades up into a densely welded, dominantly andesitic lenticulite. The Upper Tikorangi ignimbrite (UTi) is poor in lithics, crystals, and pumice, but with both andesite and rhyolite pumice clasts common up to the top of the ignimbrite.The geochemistry and petrography of vitric fragments and fiamme indicate that both magma mixing and mingling between andesite and rhyolite have occurred. Mixing occurred when hotter andesite magma was injected into st reservoir of more viscous rhyolitic magma and immediately triggered a violent eruption. The initial stage of the eruption was driven by superheating of the rhyolite magma and continued degassing. Subsequently, chamber evacuation permitted caldera block collapse that drove the eruption by pumping out the remaining magma. An increase in rifting within the Taupo Volcanic Zone, at 0.90 Ma, may have facilitated the eruption process.