Colin J. N. Wilson

Volcanic and structural evolution of Taupo Volcanic Zone, New Zealand: a review

The Taupo Volcanic Zone (TVZ) in the central North Island is the main focus of young volcanism in New Zealand. Andesitic activity started at c. 2 Ma, joined by voluminous rhyolitic (plus minor basaltic and dacitic) activity from c. 1.6 Ma. The TVZ is c. 300 km long (200 km on land) and up to 60 km wide, as defined by vent positions and caldera structural boundaries. The total volume of TVZ volcanic deposits is uncertain because a sub-volcanic basement has not been identified, but present data suggest bulk volumes of 15–20,000 km3, and that faulted metasediments form most of the immediate subvolcanic basement. Rhyolite (≥15,000 km3 bulk volume, typically 70–77% SiO2) is the dominant magma erupted in the TVZ (mostly as calderaforming ignimbrite eruptions), andesite is an order of magnitude less abundant, and basalt and dacite are minor in volume (< 100 km3 each). The history of the TVZ is here divided into ‘old TVZ’ from 2.0 Ma to 0.34 Ma, and ‘young TVZ’ from 0.34 Ma onwards, separated by the Whakamaru eruptions, which obscured much of the evidence for older activity within the zone. The TVZ shows a pronounced segmentation into northeastern and southwestern andesite-dominated extremities with composite cones and no calderas, and a central 125-km-long rhyolite-dominated segment. Eight rhyolitic caldera centres have so far been identified in the central segment, of which two (Mangakino and Kapenga) are composite features, and more centres will probably be delineated as further data accumulate. These centres account for 34 inferred caldera-forming ignimbrite eruptions, in the c. 1.6-Ma lifetime of the central TVZ. The modern central TVZ is the most frequently active and productive silicic volcanic system on Earth, erupting rhyolite at c. 0.28 m3 s−1, and available information suggests this has been so for at least the past 0.34 Ma. The rhyolites show no major compositional changes with time, though the extent of magma chamber zonation may have changed with the incoming of rifting and crustal extension in the past c. 0.9 Ma. Within the central TVZ, non-rhyolitic compositions have been erupted apparently irregularly in time and space; in particular there is no evidence for a geographic separation of basalts from andesites. Between 0.9 and 0.34 Ma, a major episode of uplift affected areas around the TVZ, while at the same time the main focus of activity may have migrated eastwards within the TVZ accompanying rifting along the axis of the zone. The modern TVZ is rifting at rates between 7 and 18 mm a−1 and restoration of the thin (15km) ‘crust’ (Vp ≤ 6.1 km s−1) beneath the central TVZ to its pre-rifting thickness (25 km) implies that rifting at such rates may have begun only at c. 0.9 Ma. The TVZ is a rifted arc, but its longitudinally segmented nature, high thermal flux and voluminous rhyolitic volcanism make it unique on Earth.

A vesicularity index for pyroclastic deposits

The vesicularity of juvenile clasts in pyroclastic deposits gives information on the relative timing of vesiculation and fragmentation, and on the role of magmatic volatiles versus external water in driving explosive eruptions. The vesicularity index and range are defined as the arithmetic mean and total spread of vesicularity values, respectively. Clast densities are measured for the 16–32 mm size fraction by water immersion techniques and converted to vesicularities using measured dense-rock equivalent densities. The techniques used are applied to four case studies involving magmas of widely varying viscosities and discharge rates: Kilauea Iki 1959 (basalt), Eifel tuff rings (basanite), Mayor Island cone-forming deposits (peralkaline rhyolite) and Taupo 1800 B.P. (calc-alkaline rhyolite). Previous theoretical studies suggested that a spectrum of clast vesicularities should be seen, depending on the magma viscosity, eruption rate, and the presence and timing of magma: water interaction. The new data are consistent with these predictions. In magmatic “dry” eruptions the vesicularity index lies uniformly in the range 70%–80% regardless of magma viscosity. For high viscosities and eruption rates the vesicularity ranges are narrow (< 25%), but broaden to between 30% and 50% as the viscosity and eruption rates are lowered and the volatiles and magma can de-couple. In phreatomagmatic “wet” eruptions, widely varying clast vesicularities reflect complex variations in the relative timing of vesiculation and water-induced fragmentation. Magma:water interaction at an early stage greatly reduces the vesicularity indices (< 40%) and broadens the ranges (as high as 80%), whereas late-stage interaction has only a minor effect on the index and broadens the range to a limited extent. Clast vesicularity represents a useful third parameter in addition to dispersal and fragmentation to characterise pyroclastic deposits.

Chronology and dynamics of a large silicic magmatic system: Central Taupo Volcanic Zone, New Zealand

The central Taupo Volcanic Zone in New Zealand is a region of intense Quaternary silicic volcanism accompanying rapid extension of continental crust. At least 34 caldera-forming ignimbrite eruptions have produced a complex sequence of relatively short-lived, nested, and/or overlapping volcanic centers over 1.6 m.y. Silicic volcanism at Taupo is similar to the Yellowstone system in size, longevity, thermal flux, and magma output rate. However, Taupo contrasts with Yellowstone in the exceptionally high frequency, but small size, of caldera-forming eruptions. This contrast reflects the thin, rifted nature of the crust, which precludes the development of long-term magmatic cycles at Taupo. 11 refs., 4 figs., 1 tab.

The Bishop Tuff: New Insights From Eruptive Stratigraphy

The 0.76 Ma Bishop Tuff, from Long Valley caldera in eastern California, consists of a widespread fall deposit and voluminous partly welded ignimbrite. The fall deposit (F), exposed over an easterly sector below and adjacent to the ignimbrite, is divided into nine units (F1‐F9), with no significant time breaks, except possibly between F8 and F9. Maximum clast sizes are compared with other deposits where accumulation rates are known or inferred to estimate an accumulation time for F1‐F8 as ca. 90 hrs. The ignimbrite (Ig) is divided into chronologically and/or geographically distinct packages of material. Earlier packages (Ig1) were emplaced mostly eastward, are wholly intraplinian (coeval with fall units F2‐F8), Lack phenocrystic pyroxenes, and contain few or no Glass Mountain‐derived rhyolite lithic fragments. Earlier packages (Ig2) were erupted mostly to the north and east, are at least partly intraplinian (interbedded with fall unit F9 to the east), contain pyroxenes, and have lithic fractions rich in Glass Mountain‐derived rhyolite or other lithologies exposed on the northern caldera rim. Recognition of the intraplinian nature of Ig1 east of the caldera and use of the fall deposit chronometry yields accumulation estimates of ca. 25 hrs for an earlier, less‐welded subpackage and ca. 36 hrs for a later, mostly welded subpackage. Average accumulation rates range up to ≥1 mm/s of dense‐welded massive ignimbrite, equivalent to ≥2.5 mm/s of non‐welded material. Comparisons of internal stratification in Ig1 and northern Ig2 lobes suggest the thickest northern ignimbrite accumulated in ≥35 hrs. Identifiable vent positions migrated from an initial site previously proposed in the south‐central part of the caldera (F1‐8, Ig1) in complex fashion; one vent set (for eastern Ig2) migrated east and north toward Glass Mountain, while another set (for northern Ig2) opened from west to east across the northern caldera margin. Vent locations for Ig1 and Ig2 southwest of the caldera have not been identified. The new stratigraphic framework shows that much of the Bishop ignimbrite is intraplinian in nature, and that fall deposits and ignimbrite units previously inferred to be sequential are largely or wholly coeval. Fundamental reassessment is therefore required of all existing models for the eruption dynamics and the nature and causes of pre‐eruptive zonations in trace elements, volatiles, and isotopes in the parental magma chamber.

The Taupo Eruption, New Zealand I. General Aspects

The ca. a.d. 186 Taupo eruption was the latest eruption at the Taupo Volcanic Centre, occurring from a vent, at Horomatangi Reefs, now submerged beneath Lake Taupo in the central North Island of New Zealand. Minor initial phreatomagmatic activity was followed by the dry vent 6 km3 Hatepe plinian outburst. Large amounts of water then entered the vent during the 2.5 km3 Hatepe phreatoplinian ash phase, eventually stopping the eruption, though large amounts of water continued to be ejected from the vent area, causing gullying of the ash deposits. After a break of several hours to weeks, phreatoplinian activity resumed, generating the 1.3 km3 Rotongaio ash, notable for its fine grainsize and for containing significant quantities of non- or poorly-vesicular juvenile material. The vent area then became dry again, and eruption rates and power markedly increased into the 23 km3 Taupo ‘ ultraplinian ’ phase, which is the most powerful plinian outburst yet documented. Synchronous with this ultraplinian activity, lesser volumes of non- to partially-welded ignimbrite were generated by diversion of ejecta from, or partial collapse of, the eruption column. The rapid rate of magma withdraw al during this phase removed support from the vent area, to trigger local vent collapse and initiate the catastrophic eruption of the ca. 30 km3 Taupo ignimbrite. Finally, after some years, lava was extruded on to the floor of the reformed Lake Taupo, and floating fragments derived from the lava carapace were driven ashore. The known eruption volume is more than 65 km3, while additional volumes are represented by primary material now beneath Lake Taupo and layer 3 to the ignim brite phases; a total volume of more than 105 km3 is likely, equivalent to more than 35 km3 of magma plus more than 3 km3 of lithic debris. Airfall deposits more than 10 cm thick blanketed 30000 km2 of land east of the vent, while ignimbrite covers a near-circular area of 20000 km2. Widespread and locally severe ground shaking occurred during, but mostly after the eruption, associated with subsidence in the Lake Taupo basin. Secondary deposits are abundant above and extending beyond the Taupo ignimbrite, consisting of the products of surface water interacting with the still-hot ignimbrite and subsequent water reworking of the light, pumiceous materials. The complexity and size of this eruption preclude accurate forecasting of the size, nature and return period of the inevitable next eruption from the Taupo Volcanic Centre.

The role of fluidization in the emplacement of pyroclastic claws: An experimental approach

Abstract A series of experiments was carried out to review the process of fluidization for a number of particulate materials having various sorting and grain shape characteristics. The up (increasing gas velocity) curve on a gas velocity/bed-pressure drop plot for a poorly sorted mixture of irregularly shaped particles is divided into three sections; non-expanded, expanded, and segregating. These sections are used to define a threefold genetic classification of pyroclastic flows which can be directly linked to conditions within a semifluidized parent flow. Type 1 flows are ungraded and mostly result from hot-avalanche flows and pyroclastic flows formed of relatively dense material. Type 2 and type 3 flows are mainly pumiceous ignimbrites and are distinguished by expansion induced coarse-tail grading, coupled with segregation structures in the latter. The implications of this classification are discussed with reference to the flow regimes, deposition slope angles, crystal concentration and “fossil fumaroles” variously developed in the flows. The relevance of the source of fluidizing gas is discussed in relation to the zoning of flow types within a single flow unit and it is shown how this, and its attendant structures, can be used to estimate the relative importance of each gas source during and after flow emplacement.

The Taupo Eruption, New Zealand II. The Taupo Ignimbrite

The ca. 30 km3 Taupo ignimbrite was erupted as a climax to the ca. AD 186 Taupo eruption in the central North Island of New Zealand. It was erupted as a single vent-generated flow unit over a time period of ca. 400 s and was emplaced very rapidly (locally at more than 250-300 m s-1) and violently. The parent flow reached 8 0 + 10 km from source in all directions, crossed all but one of the mountains within its range and only stopped when it ran out of material. The ignimbrite is divisible into layers 1 and 2, and a distant facies which combines features of both layers. Layer 1 was generated as a result of strong fluidization in the flow head, caused by air ingestion, and consists of two main facies. Layer 1(P) is a pumiceous, mildly to strongly fines-depleted unit, generated by the expulsion of material from the flow front, and termed the jetted deposits. The overlying layer 1 (H) is a thinner, crystal- and lithic-rich, fines-depleted unit, generated by the sedimentation of coarse/dense constituents segregated out by strong fluidization within the flow head and termed the ground layer. Layer 2 consists of two facies with similar compositions but contrasting morphologies; during emplacement, material left behind by the flow body partially drained into depressions to form the valley-ponded ignimbrite, leaving the veneer deposit as a thin, landscape mantling layer on interfluves. The distant facies occurs in some outermost hilly areas of the ignimbrite where the flow velocity remained high but its volume had shrunk through deposition so that air ingestion fluidization affected the whole flow. The ignimbrite shows great lateral variations. Each facies, or variants therein, exhibits systematic degrees of development with varying distances from vent. Near vent, the flow consisted of a series of batches of material which by ca. 25 km had coalesced into a single wavy flow and by ca. 40 km into a single wave. Out to ca. 13 km, the flow was rather dilute and highly turbulent as it deflated from the collapsing eruption column. Beyond this distance it was fairly concentrated, being less than 100% expanded over its non-fluidized compacted state, and had acquired a fluidization-induced stable density stratification, which strongly suppressed turbulence in the flow body. Deflation from the eruption column was largely complete by ca. 13 km but influenced the flow as far as 20-25 km from vent. Grainsize and compositional parameters measured in the ignimbrite show lateral variations which equal or exceed the entire spectrum of published ignimbrite data. The flow had deflated and coalesced from the eruption column by ca. 20 km from vent. Beyond this distance most lateral variations are modelled by considering the flow to be a giant fluidized bed. As the flow moved, material was deposited from its base, and hence predictable vertical variations in the model fluidized bed are comparable with lateral variations in the ignimbrite. The agreement is excellent, and, in particular, discontinuities in the nature of the ignimbrite at 55-60 km from vent suggest that the more distal ignimbrite represents a vast segregation layer generated above the moving flow. Differences between the model and variations of some parameters reflect the influence of kinetic processes, such as shearing and local fluidization, that operated regardless of the bulk flow composition. The strong fluidization in the flow is a result of the high flow velocities (promoting air ingestion), not vice versa as is often accepted. Contrasts in the natures of layers 1 and 2 imply that the first material erupted contained significantly coarser, and a higher content of, lithics than the bulk of the flow. During emplacement, this earlier material was depleted by deposition and diluted by material introduced from the flow body. Systematic regional variations also occur in the ignimbrite: for example, it contains lower crystal: lithic ratios and higher density pumice in a northeasterly sector, and vice versa to the southwest. Ignimbrite found in mountainous areas shows changes consistent with its derivation from the upper, more mobile and pumiceous top of the flow. Fluidization processes generated structures and facies in the ignimbrite on various scales. Individual segregation bodies found at any exposure show features mimicking those of the ground layer, i.e. fines depletion and crystal- and lithic-enrichment. Fluidization-induced grading visible at individual exposures accounts for the great range of grading styles seen in the valley-ponded ignimbrite, and strong fluidization has locally generated an upper fines- and pumice- rich segregation layer (here termed layer 2c). On the largest scale, fluidization was primarily responsible for the generation of the layer 1 deposits, and for the grainsize and compositional zonation within the flow that produced the lateral variations in the ignimbrite. Ingested and heated air is inferred to have been the most important gas source for fluidization within the flow, although several other gas sources were locally dominant. It is clear that the thickness, grainsize and composition of the ignimbrite at any point are not simply related to values of these parameters in either the originally erupted material or the parent flow, and that, except for its density, the dimensions and composition of the parent flow cannot be directly inferred from the ignimbrite.

Ignimbrite depositional facies: the anatomy of a pyroclastic flow

A model is presented for the depositional regimes in a pyroclastic flow, to explain the origin of facies in the Taupo ignimbrite and to compare these with other published examples. In this model, a pyroclastic flow consists of a head, a body and a tail. The head, which is where fluidization caused by air ingestion occurs, generates layer 1 deposits. On lithological grounds these are divided into layers 1(P) which is fines-depleted and rich in pumice, and 1(H) which is fines-depleted and rich in lithics and crystals. Layer 1(P) represents material thrown forwards from the flow head and is termed the jetted deposits. Layer 1(H) represents material sedimented from within the flow head and is termed the ground layer. The body represents the bulk of the flow and the tail is its trailing part which is slowed by ground friction; these parts generate the layer 2 deposits which include the valley-pond ignimbrite and its associated ignimbrite veneer deposit. Localized depositional modes within the body and tail generate distinctive coarse pumice concentration zones and pumiceous lee-side lenses behind obstacles. At its outer limits all of the flow on interfluves is affected by air-ingestion fluidization, producing the distant facies which combines features of both layer 1 and 2 deposits. One important conclusion of this work is that the thicknesses and compositions of the various facies are not related in any simple way to the thickness and composition of the parent flow.

The role of fluidization in the emplacement of pyroclastic flows, 2: Experimental results and their interpretation

Abstract Although often quoted as being important in the emplacement of pyroclastic flows, little relevant experimental fluidization work has been done. Fluidization processes are first reviewed for a simple system, against which ignimbrite samples can be compared. Results from fluidization experiments on ignimbrite samples show that their behaviour differs radically from any simple system, principally because of their extremely poor sorting. At a certain gas velocity, U ie (whose value cannot be predicted), some ignimbrite samples begin to expand, whilst at a higher gas velocity, U mp , the samples begin to show segregation structures. The minimum fluidization velocity, U mf , in simple systems is replaced by U mp ; the value of the latter in pyroclastic materials cannot be easily or reliably predicted from published U mf correlations, and a new empirical method for determining U mp is presented from the experimental data. During fluidization, ignimbrite samples expand much less than more conventional materials due to the bypassing of gas through segregation channels. The total amount of expansion is also limited; e.g. a 100-m-thick pyroclastic flow will, from fairly high gas velocities, deflate to form an ignimbrite which is not less than 70–85 m thick. During fluidization, the poor sorting causes only part of the weight of the ignimbrite samples to be supported by the gas flow, the degree of support increasing with an improvement in sorting. Segregation structures, and grain-size and compositional variations within fluidized ignimbrite samples show varying behaviour, depending on the sorting of the sample. After subjecting a sample to high gas velocities, the net result is a grain-size- and compositionally-zoned bed which becomes richer upwards in pumice and fine material. Once formed, these features cannot be destroyed under laboratory conditions, implying that some segregation structures should be present in any ignimbrite where U mp was exceeded.

Crystallisation ages in coeval silicic magma bodies: 238U-230Th disequilibrium evidence from the Rotoiti and earthquake flat eruption deposits, Taupo volcanic zone, New Zealand

Abstract The timescales over which moderate to large bodies of silicic magma are generated and stored are addressed here by studies of two geographically adjacent, successive eruption deposits in the Taupo Volcanic Zone, New Zealand. The earlier, caldera-forming Rotoiti eruption (>100 km3 magma) at Okataina volcano was followed, within months at most, by the Earthquake Flat eruption (∼10 km3 magma) from nearby Kapenga volcano; both generated non-welded ignimbrite and coeval widespread fall deposits. The Rotoiti and Earthquake Flat deposits are both crystal-rich high-silica rhyolites, with sparse glass-bearing granitoid fragments also occurring in Rotoiti lag breccias generated during caldera collapse. Here we report 238U–230Th disequilibrium data on whole rocks and mineral separates from representative Rotoiti and Earthquake Flat pumices and the co-eruptive Rotoiti granitoid fragments using TIMS and in situ zircon analyses by SIMS. Multiple-grain zircon-controlled crystallisation ages measured by TIMS from the Rotoiti pumice range from 69±3 ka ( 350 ka, with a pronounced peak at 70–90 ka. The weighted mean of isochrons is 83±14 ka, in accord with the TIMS data. One glass-bearing Rotoiti granitoid clast yielded an age of 57±8 ka by TIMS (controlled by Th-rich phases that, however, are not apparently present in the juvenile pumices). Another glass-bearing Rotoiti granitoid yielded SIMS zircon model ages peaking at 60–90 ka, having a similar age distribution to the pumice. Age data from pumices are consistent with a published 64±4 ka eruptive age (now modified to 62±2 ka), but chemical and/or mineralogical data imply that the granitoid lithics are not largely crystalline Rotoiti rhyolite, but instead represent contemporaneous partly molten intrusions reflecting different sources in their chemistries and mineralogies. Similarly, although the Earthquake Flat eruption immediately followed (and probably was triggered by) the Rotoiti event, age data from juvenile material are significantly different. A multiple-grain zircon-controlled crystallisation age measured by TIMS from a representative pumice is 173±5 ka, while SIMS model ages range from 70−26+34 ka to >350 ka, with a peak at 105 ka. These age data coupled with previously published geochemical and isotopic data show that the Rotoiti and Earthquake Flat deposits were erupted from independent, unconnected magma bodies.