A zircon U-Pb geochronology for the Rotokawa geothermal system, New Zealand, with implications for Taupō Volcanic Zone evolution

Journal Pre-proof A zircon U-Pb geochronology for the Rotokawa geothermal system, New Zealand, with implications for Taupō Volcanic Zone evolution

S.D. Milicich, I. Chambefort, C.J.N. Wilson, S. Alcaraz, T.R. Ireland, C. Bardsley, M.P. Simpson PII:

S0377-0273(19)30460-3

DOI:

https://doi.org/10.1016/j.jvolgeores.2019.106729

Reference:

VOLGEO 106729

To appear in:

Journal of Volcanology and Geothermal Research

Received date:

6 September 2019

Revised date:

21 October 2019

Accepted date:

13 November 2019

Please cite this article as: S.D. Milicich, I. Chambefort, C.J.N. Wilson, et al., A zircon UPb geochronology for the Rotokawa geothermal system, New Zealand, with implications for Taupō Volcanic Zone evolution, Journal of Volcanology and Geothermal Research(2019), https://doi.org/10.1016/j.jvolgeores.2019.106729

This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition of a cover page and metadata, and formatting for readability, but it is not yet the definitive version of record. This version will undergo additional copyediting, typesetting and review before it is published in its final form, but we are providing this version to give early visibility of the article. Please note that, during the production process, errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

© 2019 Published by Elsevier.

Journal Pre-proof A zircon U-Pb geochronology for the Rotokawa geothermal system, New Zealand, with implications for Taupō Volcanic Zone evolution S.D. Milicicha, I. Chambefortb, C.J.N. Wilsonc, S. Alcarazb, T.R. Irelandd, C. Bardsleye, M.P. Simpsonb a

GNS Science, Avalon Research Centre, Lower Hutt 5010, New Zealand

b

GNS Science, Wairakei Research Centre, Taupō 3352, New Zealand

c

School of Geography, Environment and Earth Sciences, Victoria University, Wellington 6140, New Zealand Research School of Earth Sciences, Australian National University, Canberra, ACT 0200, Australia

e

Mercury NZ Ltd., PO Box 245, Rotorua 3040, New Zealand

Jo ur

na

lP

re

-p

ro

of

d

Corresponding author: Tel.: +64 4 570 4879; fax: +64 4 570 4600 E-mail address: [email protected] (S.D. Milicich) Manuscript for: Journal of Volcanology and Geothermal Research Version: 13 November 2019

1

Journal Pre-proof Abstract A U-Pb zircon geochronology for the Rotokawa geothermal system, central Taupō Volcanic Zone (TVZ), New Zealand, provides new constraints on the chronostratigraphy and volcanic and structural evolution of this area and the broader TVZ. The 3-km-thick volcanic sequence at Rotokawa is mainly composed of rhyolitic ignimbrites linked to large caldera-forming events from sources outside the field area, but locally sourced andesite and rhyolite lavas and intrusions are also present. Crystallisation age spectra (and consequent estimates of eruption age) have been obtained

of

on zircons from hydrothermally altered magmatic rocks by Secondary Ion Mass Spectrometry techniques using a SHRIMP-RG instrument. The oldest rock dated is a Tahorakuri Formation

ro

ignimbrite (eruption age estimate of 1.87±0.03 Ma) which, along with comparable-age units at other

-p

nearby TVZ geothermal systems (Ngatamariki, Ohaaki), is among the oldest silicic volcanic deposits

re

known in the TVZ. These ignimbrites collectively onlap a basal andesite lava pile, up to 1.2 km thick at Rotokawa, that in turn rests on the Mesozoic basement greywacke. The base of the lava pile

lP

is more faulted than its top surface, implying that rifting and graben formation had started along the

na

line of the modern TVZ arc prior to 1.84 Ma and is not a younger feature. Between ~1.8 Ma and 700 ka, there are no rocks represented at Rotokawa, with the next oldest lithology being a 720 ± 90 ka

Jo ur

rhyolite lava. At 350 ka, the Rotokawa area was buried by regionally extensive ignimbrites of the Whakamaru Group, which have since subsided by ~700 m but not been greatly faulted. Ignimbrites and sediments of the Waiora Formation were then emplaced, coevally with widespread and volumetrically greater volcanism in the Maroa and Ngatamariki areas 13 km northwest and 8 km north of Rotokawa, respectively. Local rhyolites of the Oruahineawe Formation, dated at ~100 ka, were emplaced both as extrusive domes and shallow intrusions below the area. Sedimentary rocks of the Huka Falls Formation and deposits of the 14C-dated 25.4 ± 0.2 ka Oruanui eruption capped and sealed the system, which has since been disrupted by hydrothermal eruption events. The largest of these occurred at ~6.8 ka (14C date) broadly coincident with a resumption of eruptive activity at Taupō volcano, 20 km to the south-southwest. Notable aspects of the evolution of the Rotokawa area are the early onset of rifting and subsidence along the line of the modern arc, the lack of volcanic activity for > 1 Myr from 1.84 Ma to 720 ka, the lack of faulting and only modest subsidence since 2

Journal Pre-proof 350 ka, and the contrasts in volcanic and subsidence histories with other, nearby geothermal systems.

1. Introduction The central segment of the Taupō Volcanic Zone (TVZ) in the central North Island of New Zealand is one of the world’s most vigorously active regions of Quaternary silicic magmatism, with associated volcanism and geothermal activity (Fig. 1). The overall setting is that of a rifting arc,

of

associated with subduction of the Pacific Plate beneath the Australian Plate (Cole and Lewis, 1981; Seebeck et al., 2014a) but, within this central segment, there is an overall magmatic flux and

ro

associated geothermal heat flow about an order of magnitude greater than in a typical arc (Hochstein,

-p

1995; Wilson and Rowland, 2016, for overview). In its ~2 Myr longevity, areal extent, dominance of

re

voluminous rhyolitic volcanism and total geothermal flux (4.2 ± 0.5 GW: Bibby et al., 1995), the central TVZ closely parallels the Yellowstone system (Christiansen, 2001; Hurwitz and Lowenstern,

lP

2014). There are two marked contrasts with Yellowstone, however. First, in the central TVZ

na

somewhat less of the overall thermal flux (c. 75 %) is released by crustal heat flow, mostly at 23 high-temperature (>250 ºC) geothermal systems, many of which have been explored by drilling and

Jo ur

utilised for power generation (Bibby et al., 1995; Kissling and Weir, 2005; Chambefort and Bignall, 2016). Second, in the central TVZ a higher proportion of the overall thermal flux reaches the surface in the form of more numerous, but generally smaller eruptions (Houghton et al., 1995; Wilson et al., 2009). Deposits of these eruptions are proven from geothermal drilling to have collectively accumulated to >3 km thicknesses over much of the central TVZ in areas down-dropped by caldera collapse or rifting. The drilling programmes associated with the exploration and development of geothermal power projects provide unique insights into the subsurface architecture of the TVZ rift infill. The deepest wells (up to 3.5 km) allow the stratigraphy of the individual geothermal systems to be investigated, in some cases down to the pre-volcanic Mesozoic greywacke basement. The systemspecific correlation and reconstruction of stratigraphy is vital to unravelling the complex volcanic and structural history of the host rocks to the geothermal systems and linking these localised areas 3

Journal Pre-proof with the wider central TVZ environment. However, stratigraphic correlation is often challenging due to strong hydrothermal alteration that can destroy the original igneous minerals and textures, and occasionally severe mixing of drill cuttings that hinders accurate correlations (Milicich et al., 2013a). In general, only limited correlations can be made with dated surficial lithologies through petrographic characteristics alone (e.g., crystal species, abundances and sizes). Strong hydrothermal alteration of igneous rocks hosting the geothermal systems precludes the use of most radiometric dating applications (e.g., Ar-Ar on plagioclase). However, zircon, a trace mineral present in many

of

igneous rocks in the TVZ, is highly resistant to hydrothermal alteration processes and can be dated by U-Pb techniques to provide limits on the ages (crystallisation, intrusion, or eruption) of the host

ro

lithology (Dalrymple et al., 1999; Schmitt et al., 2003). In the TVZ context, zircon age dating has

-p

proved invaluable at reconstructing field histories and correlating major regional ignimbrite units

re

(Wilson et al., 2008, 2010; Milicich et al., 2013a; Eastwood et al., 2013; Chambefort et al., 2014; Rosenberg et al., 2019). Here we present a suite of zircon U-Pb age estimates from altered

lP

subsurface rocks from the Rotokawa geothermal system and combine these data with a re-

it in the broader TVZ context.

na

assessment of the subsurface stratigraphy to reconstruct the geological history of the area and place

Jo ur

Rotokawa is one of the highest temperature geothermal systems in the TVZ, with a maximum measured temperature of 337°C (well RK22: Sewell et al., 2015). The Rotokawa wells collectively record a complete stratigraphic succession (i.e., the deepest wells intersect Mesozoic basement greywacke) and document a transition from earlier andesitic to later rhyolitic volcanism (Browne et al., 1992). The onset of the latter is widely linked with the development of regional extensional tectonics and caldera development within the central TVZ (Cole, 1990; Wilson et al., 1995; Rowland et al., 2010; Deering et al., 2011), although more recent work (Eastwood et al., 2013; Chambefort et al., 2014) has suggested that this transition occurred at approximately 1.8 Ma, earlier than previously proposed. Here we document stratigraphic information from drillholes through the Rotokawa geothermal system derived from our detailed lithological review of drillcore and cuttings from the geothermal wells and couple this knowledge with new U-Pb age data on zircons. Following earlier practise (e.g., Milicich et al., 2013a), we use the term ‘Rotokawa geothermal system’ to refer 4

Journal Pre-proof to the natural entity, and ‘Rotokawa Geothermal Field’ to refer to the resource delineated and utilised by drilling.

2. Rotokawa geothermal system 2.1. Hydrothermal setting The Rotokawa geothermal system is located in the southern part of the central TVZ, about 10 km NE of Taupō township (Fig. 1). Drilling and geophysical investigations since the 1960s have

of

identified a large high-temperature (>300 °C) resource with an extent of ~28 km2, as delineated by the 30 Ωm resistivity contour (Cole and Legmann, 1998; Risk, 2000). There are numerous surficial

ro

thermal manifestations in the Rotokawa geothermal system (Fig. 2), mostly concentrated in two

-p

areas (Krupp and Seward, 1987; Brotheridge, 1995; Milicich and Hunt, 2007; Price et al., 2011). In a

re

southern area, the acid–sulfate (pH ~2) Lake Rotokawa (Fig. 2) occupies a hydrothermal eruption crater (Browne and Lawless, 2001; Collar and Browne, 1985). This area also includes springs in a

lP

steam-heated zone referred to as ‘the lagoon’ on the NE shore of Lake Rotokawa, a fumarole, some

na

acid–sulfate–chloride springs along the Parariki Stream, and a silica-rich, acid fluid discharge actively depositing sinter on the flood plain of the Parariki Stream (Schinteie et al., 2007). A

Jo ur

northern area is represented by relatively fewer thermal features including chloride–bicarbonate springs located on the banks of the Waikato River, and a small area of steaming ground near well RK8 (Fig. 2).

Rotokawa is a gas-rich high-temperature geothermal system (Giggenbach, 1995) with three aquifers: (a) shallow meteoric groundwater; (b) a complex intermediate aquifer with several different fluid types including steam heated groundwater, acid-sulfate-chloride fluids and boiled reservoir (chloride) fluids; and (c) a chloride geothermal reservoir of >300 °C below 1000 m depth (Winick et al., 2011; Addison et al., 2015). Rotokawa hosts some of the highest contents of H2S and CO2 amongst the central TVZ geothermal systems, with estimated emissions of 441 t d-1 of CO2 and up to 31 t d-1 of H2S (Bloomberg et al., 2014). The high H2S and CO2 contents of the reservoir fluid coupled with boiling at shallow levels and mixing with shallow aquifer groundwater yields acidsulfate and bicarbonate fluids. As a result, hydrothermal alteration at the surface, is primarily driven 5

Journal Pre-proof by these acid fluids, depositing large amounts of native sulfur in addition to kaolinite, smectite, silica, alunite and other sulfates, cinnabar, and arsenic precipitates (Chambefort et al., 2011; Price et al., 2011). Sulfate minerals are identified also to depths of ~1600 m (Hedenquist et al., 1988; Chambefort et al., 2011). The main alteration assemblage in the reservoir is propylitic, generated in the host rocks by high-temperature, near-neutral pH chloride fluids. The alteration mineralogy includes quartz, chlorite, illite, calcite, pyrite, adularia, albite, epidote, clinozoisite, wairakite, and accessory hematite (Krupp et al., 1986; Krupp and Seward, 1987; Price et al., 2011). This mineral

of

association overprints a higher temperature assemblage which includes actinolite and biotite (Chambefort et al., 2011). The timing of this higher temperature alteration cannot however be linked

-p

ro

to any specific magmatic or volcanic event that we document below.

re

2.2. Surface and subsurface geology

In the north of the Rotokawa Geothermal Field, the surface geology is dominated by the

lP

coalesced Oruahineawe and Kaimanawa rhyolite domes (Fig. 2; Leonard et al., 2010; Downs et al.,

na

2014a). To the east, there is the deeply eroded cone of the 712 ± 27 ka (all errors are 2 s.d. unless otherwise stated) Rolles Peak andesite (K-Ar age: Tanaka et al., 1996), and to the west there is a 103

Jo ur

± 12 ka rhyolite lava exposed at Aratiatia dam (Ar-Ar age: Downs et al., 2014a). Elsewhere, the surface geology is composed of young pyroclastic rocks and sediments. The 232 ± 10 CE Taupo ignimbrite and alluvium derived from it form terraces near the Waikato River, and exposures of the 25.4 ± 0.2 ka Oruanui ignimbrite can be found between the Waikato River and Lake Rotokawa (Manville et al., 1999; Leonard et al., 2010; Hogg et al., 2012; Vandergoes et al., 2013). The area around Lake Rotokawa is blanketed by hydrothermal eruption breccias and Holocene tephras (Collar and Browne, 1985; Browne and Lawless, 2001). The Lake Rotokawa hydrothermal eruption crater, together with at least twelve other hydrothermal eruption craters and associated eruption deposits, represents local hydrothermal activity since 25.4 ka (Collar and Browne, 1985; Krupp and Seward. 1987). Hydrothermal eruption craters have mainly been infilled by younger tephras and sediments and are located by mapping clast size variations in their inferred breccia ejecta deposits (Browne and Lawless, 2001). The Parariki Breccia succession, composed of several breccia deposits erupted from 6

Journal Pre-proof depths of up to 450 m, cover much of the south of the field (Collar and Browne, 1985). Breccia from the largest hydrothermal eruption (6,060 ± 120 years ago, uncalibrated Libby 14C age: Collar, 1985), inferred to have originated from the Lake Rotokawa area, covers an area of ∼12 km2 to a maximum exposed thickness of ∼11 m (Collar and Browne, 1985; Browne and Lawless, 2001). As in all of the drilled geothermal systems within the TVZ, correlating individual eruptive and sedimentary units within the sub-surface silicic volcanic sequence at Rotokawa is challenging. This is due in part to the masking of primary rock textures by hydrothermal alteration and also in

of

part to the paucity of drillcore recovery in modern drilling practices and the associated fine grain-

ro

size of available drilling cuttings (Milicich et al., 2013a, Chambefort et al., 2014). Additionally, in the Rotokawa wells there are substantial vertical intervals where no drill cuttings were returned to

-p

surface (i.e., blind drilling), or where there was significant mixing of cuttings and identification of

re

geological units and their contacts is unreliable. For this reason, the geological succession at Rotokawa has been re-considered and a new presentation of the geological units is given in section 4

lP

and summarised in Table 1, Figure 3, and Online Appendix 2. This stratigraphic re-examination

na

helped shape the choice of materials to be dated, which was also constrained by whether or not zircons were available. For this study, zircons were extracted from eleven drillcore samples from

Jo ur

different Rotokawa formations for single-crystal U-Pb dating by Secondary Ion Mass Spectrometry (SIMS) and added to the single age determination reported by Eastwood et al. (2013). A summary of the dating samples is included in Table 1, with the full data set in Online Appendix 1.

3. Methods Drillcore samples selected for zircon extraction were ground in a ringmill and sieved to yield a <250m size fraction. Concentration of the heavy mineral fraction was achieved by density separation using sodium polytungstate. The concentrates were rinsed in warm nitric acid to remove the often-high amounts of pyrite, followed by further density separation with methylene iodide. Due to the presence of other dense alteration minerals (often titanite and pyrite), zircons were handpicked from a final concentrate on the basis of their distinctive morphologies and adamantine lustre.

7

Journal Pre-proof Grains were mounted in 25 mm epoxy resin pucks, polished to approximately mid-way through the grains, then cathodoluminescence (CL) imaged using a Robinson Detector on a JEOL 6610 Scanning Electron Microscope at the Research School of Earth Sciences (RSES), Australian National University. Age determinations for this study were acquired using the Sensitive HighResolution Ion MicroProbe–Reverse Geometry (SHRIMP-RG) instrument at the RSES. In order to minimise contamination by common Pb, mounts were rinsed in detergent and HCl prior to gold coating and prior to both the CL imaging and ion probe analysis sessions. Before

of

analysis on the SHRIMP-RG, the primary beam was rastered for 180 s on a 35 x 45 µm area to remove the gold coat and any possible surface contamination. Ions were sputtered from the zircons

ro

with a 3-4 nA primary O2 beam focussed to a ~25 × 35 µm spot. The mass spectrometer was cycled

-p

through peaks corresponding to 90Zr216O, 204Pb, background, 206Pb, 207Pb, 208Pb, 238U, 232Th16O and U16O, with a total analysis time of ~900 s. Extended count times were used for background (20 s),

re

238

206

Pb (40 s) and 207Pb (30 s) and six scans were run through the mass sequence. The concentration

lP

standard used was SL13 (238 ppm U) and the age standard was R33 (420 Ma: ID-TIMS age from:

na

Zeh et al., 2015). Data reduction was carried out using SQUID 2 (Version 1.51: Ludwig, 2009). The initial 230Th deficit was corrected for by using the measured Th and U concentrations

Jo ur

(relative to SL-13), and a whole-rock Th/U value of 4.4 (following standard practise for TVZ geothermal samples; Wilson et al., 2010). The presence of common Pb was monitored by using 204

Pb, and a correction applied using the measured 207Pb/206Pb values for the analytical spot and a

common Pb isotopic composition of 207Pb/206Pb = 0.836 from the average crustal value of Stacey and Kramers (1975). Samples vary widely in the proportions of 206Pb attributable to common Pb, and various cut-offs were chosen (see Online Appendix 1) to exclude analyses with plausible, but imprecise, ages, yet leave remaining sufficient data to provide a reasonable estimate of the crystallisation age (as an approximate equivalent to the eruptive age). Older samples overall had lower proportions of common Pb. High proportions of 206Pb attributable to common Pb are not considered to be solely the result of hydrothermal alteration of the zircon structure, as the crystals do not display breakdown or replacement textures seen in altered grains elsewhere (see Milicich et al.,

8

Journal Pre-proof 2013a), but in part also reflect the generally low proportions of radiogenic Pb present in these young rocks.

4. Subsurface geology of the Rotokawa Geothermal Field In this section, a new overview is given of the geological succession in the Rotokawa Geothermal Field, based on comprehensive re-examination of the available drillcores and cuttings. A comparison of the new and previous stratigraphies is presented in Fig. 3, with cross sections

of

compared in Fig. 4. Units termed ‘lava’ are generally inferred to be surficial lava flows, but in some cases have been demonstrated from contact relationships and/or age dating contrasts with their host

ro

rocks to be intrusive (e.g., Oruahineawe Formation: this study, and the Caxton rhyolite bodies at

re

-p

Kawerau: Milicich et al., 2013a).

4.1. Torlesse Supergroup

lP

The basement rocks at Rotokawa are part of the Triassic-Early Cretaceous metasedimentary

na

sequences of indurated sandstone and mudstone of the Torlesse Supergroup (Figs. 3 and 4A,B; Mortimer, 1995; Adams et al., 2009; Ring et al., 2019). At Rotokawa, the greywacke basement has

Jo ur

been reached by 9 wells (wells RK4, RK16, RK19ST, and RK19-24; Fig 1) between approximately 1630 and -1960 masl (metres above sea level). The rocks are weakly to moderately hydrothermally altered interbedded grey, medium to fine grained litharenite sandstone (greywacke) and minor argillite. Detrital quartz, plagioclase, K-feldspar, biotite, and volcanic rock fragments are the main constituents enclosed in an abundant (31%) matrix. With the exception of metamorphic veins with REE-bearing epidote, monazite, pyrite, chlorite and quartz mineral association, veins of quartz, chlorite, illite with minor epidote and calcite, present in all samples examined, cannot be linked to any specific alteration timescale.

4.2. Reporoa Group The term Reporoa Group (Gravley et al., 2006) encompasses all volcanic and sedimentary stratigraphic units drilled between the greywacke basement and the 350-340 ka Whakamaru Group 9

Journal Pre-proof ignimbrites. The Reporoa Group includes the Tahorakuri (volcaniclastic materials) and Waikora (greywacke sediment) formations, together with individually named local units. At Rotokawa, the Reporoa Group is dominated by andesitic lavas and breccias, but also includes materials representative of the two clastic formations (Fig. 4). Reporoa Group andesites. A thick sequence (up to 1.6 km in RK16) of Reporoa Group andesite lavas and breccias (Browne et al., 1992; Figs. 3 and 4C) are a lower andesite sequence, here formally named Rotokawa Andesite, and the shallower andesite formally named the Nga Awa Purua

of

Andesite (cf. Wallis et al., 2013). The Rotokawa Andesite has a proposed type section in RK16 (-1130 to -2740 masl). It is

ro

crystal-rich (20-30 vol. % of the rock), with primary phenocrysts of plagioclase and clinopyroxene,

-p

plus subordinate (<5 vol.% of the rock) amphibole (Fig. 5) and accessory magnetite. The Rotokawa

re

Andesite lies directly on the greywacke basement surface and is inferred to represent the collective products of a substantial volcano (comparable in volume to the modern Ruapehu edifice, in the

lP

southern TVZ: Hackett and Houghton, 1989; Price et al., 2012; Conway et al., 2016, 2018). Drill

na

core of the andesite shows it to have a remarkably consistent porphyritic texture, commonly with chlorite-filled amygdales and only rarely exhibiting breccia textures. The uniformity of the andesite

Jo ur

is challenging to determine, however, with confidence from small cuttings that are moderately to strongly hydrothermally altered. One core sample from well RK24-ST1 (-1401 masl) incorporates a hyaloclastic breccia (Fig. 5D) indicating that some of the andesite was emplaced near or under water. Primary pyroxene phenocrysts in the Rotokawa Andesite are partially to fully replaced by calcite, titanite, chlorite and rarely anhydrite; plagioclase is often replaced by albite, adularia and calcite while the groundmass is altered and vesicles filled by chlorite and quartz. Pyrite, secondary iron and titanium oxides and titanite are also rarely present. Quartz, calcite, epidote with some anhydrite and chalcedony veins are rare to common. The shallower Nga Awa Purua Andesite (Fig. 4) was first named in Wallis et al. (2013), and the name is formalised here, with a type section proposed between -537 to -1046 masl in RK27. It is petrographically distinct from the Rotokawa Andesite in having amphibole phenocrysts (up to 5 mm long; Fig. 5E), together with plagioclase, pyroxene and (in places) minor biotite, and has a high 10

Journal Pre-proof crystal content of ~45 vol.%. There are intervals of brecciation (Fig. 5D), as well as intervals of ashgrade andesite material. The known extent of Nga Awa Purua Andesite is only in the southwest part of the Rotokawa field. The primary mineralogy and texture are partially obscured by hydrothermal chlorite, epidote, calcite, illite, anatase, and quartz. Price et al. (2011) reports rare biotite phenocrysts that are completely altered to minor actinolite-tremolite, green hydrothermal biotite, leucoxene and calcite. Tahorakuri Formation. At Rotokawa the Tahorakuri Formation consists of ~50 to 320 m of

of

silicic pyroclastic and volcaniclastic rocks (Fig. 3). The variable thickness reflect deposition over relief created by prior faulting of the underlying Rotokawa Andesite, with grabens creating localised

ro

depocenters. The formation is dominated by weakly welded pumiceous lithic- and crystal-rich tuffs

-p

(ignimbrite). Lithic clasts include andesite, rhyolite lavas, and greywacke; crystals are quartz, and

re

variably altered plagioclase and ferromagnesian minerals (Figs. 3 and 4F). The primary volcanic

chlorite and hematite.

lP

textures in the tuff matrices are masked by intense hydrothermal alteration to quartz, illite, calcite,

na

Waikora Formation. The Waikora Formation was first described by Grindley and Browne (1968) as intervals in drillholes where greywacke pebble conglomerate is dominant over tuffaceous

Jo ur

material. The Waikora Formation at Rotokawa consists of a strongly illite and chlorite altered greywacke and argillite pebble conglomerate (Fig. 5G). The Waikora Formation is intercalated with silicic pyroclastic deposits of the Tahorakuri Formation and attains its greatest thickness (245 m) in RK6 where it is inferred to fill a graben in the Rotokawa Andesite (Fig. 4). Phenocryst-poor rhyolite lava. In the north of the Rotokawa Geothermal Field one drillhole (RK8) intercepts a rhyolitic lava up to 380 m thick (Fig. 4), representing effusive volcanism at Rotokawa that occurred between the deposition of Tahorakuri Formation ignimbrites and the Whakamaru Group ignimbrite. It is not separately named here, due to its limited known extent. The rhyolite is phenocryst poor (variably 5% to < 1 vol%), with variably altered plagioclase (< 1.5 mm length), and very rare quartz and chlorite-pseudomorphed ferromagnesian minerals.

4.3. Whakamaru Group 11

Journal Pre-proof An ignimbrite unit, ∼200–390 m thick, covers much of the Tahorakuri Formation across the field, absent (within the limitations imposed by sampling intervals) only where high-standing bodies of andesite or rhyolite lava were present at the time of deposition (Fig. 4). Based on its stratigraphic position, crystal assemblage and crystal abundance, it has been correlated with ignimbrites of the Whakamaru Group (originally logged in the Wairakei and Ohaaki geothermal fields as Wairakei or Rangitaiki ignimbrites; e.g., Grindley, 1965; Grindley and Browne, 1968). These ignimbrites were erupted at 349 ± 4 ka in association with formation of a large caldera, the eastern rim of which has

of

been proposed to lie just west of Rotokawa (Wilson et al., 1986; Houghton et al., 1995; Leonard et

ro

al., 2010; Downs et al., 2014b). The slightly younger (339 ± 5 ka: Downs et al., 2014b), closely related Paeroa Subgroup ignimbrites, inferred to be present at Ngatamariki (Chambefort et al.,

-p

2014), have not been identified at Rotokawa. At Rotokawa the Whakamaru Group ignimbrite is

re

partially-welded, and contains abundant anhedral, crystal fragments (25%) of embayed quartz (<9 mm across), variably altered plagioclase (<3 mm in length) and rare to minor pseudomorphed biotite

lP

crystals and other ferromagnesian minerals (Fig. 5H). Lithic clasts are rounded to angular

na

(abundance 5%; <40 mm across) and are porphyritic and spherulitic rhyolite lava, siltstone, porphyritic andesite lava, and plutonic rock fragments. Pumice clasts are common and have been

4.4. Huka Group

Jo ur

completely altered to clays and quartz.

The Huka Group includes all stratigraphic units occurring between the Whakamaru Group and the 25.4 ka Oruanui Formation (Vandergoes et al., 2013) and records c. 300 kyr of volcanism, and volcaniclastic and sedimentary deposition. This group was deposited over a time period when silicic volcanism and extensional rifting combined to develop a complex interplay of lacustrine and subaerial depositional environments through the central TVZ (Leonard, 2003; Manville and Wilson, 2004; Downs et al., 2014a; Cattell et al., 2016) including the Rotokawa area. Waiora Formation. The Waiora Formation (Fig. 4) is a mixed volcanic and sedimentary succession, including rhyolitic lava flows, between the Whakamaru Group and the overlying finegrained lacustrine sediments of the Huka Falls Formation (Rosenberg et al., 2009). Grindley (1965) 12

Journal Pre-proof identified five members in this formation in the Wairakei Geothermal Field. In the Tauhara Geothermal Field, ~5 km SW of Rotokawa, it is cumulatively 2500 m thick (Rosenberg et al., 2019). In contrast at Rotokawa, this formation is thinner (<390 m) and the distinct members cannot be identified. At Rotokawa, the Waiora Formation mostly consists of a 20 to 390 m thick, poorly crosscorrelated succession of pumice-rich, crystal-poor, non-welded ignimbrite (Fig. 5I) and lapilli tuff units interstratified with tuffaceous coarse sandstone and siltstone layers of varied thickness.

of

Rhyolitic and andesitic lithics are common in the pyroclastic layers. In some wells (e.g., RK6), the formation is dominated by interbedded sandstone and siltstone, with only minor tuff units. In the SE

ro

of the field around RK22 and RK24, there is a ~45 m thick bed of tuffaceous siltstone. Greywacke

re

Waiora Formation and younger deposits.

-p

conglomerates, represented by the pre-350 ka Waikora Formation, are notably absent from the

There are three other lithologies present in the Waiora Formation at Rotokawa. First, basalt

lP

(drilled by wells RK16, RK17, RK18 and RK27 between ~-340 and -520 masl) occurs as scoria

na

lapilli tuff interbedded with rhyolitic tuffs. The basalt scoria are dark grey and vesicular, porphyritic to trachytic with unusually fresh phenocrysts of plagioclase (An73-60, Chambefort unpublished data),

Jo ur

pyroxene and olivine (Fig 4J). Directly underlying the basalt is a diatomaceous tuff (Fig. 5K), which indicates that the basalt was emplaced into a lake. Second, is an andesitic lapilli tuff drilled in RK6 from -401 to -437 masl. It is characterised by sub- to well-rounded clasts of highly vesicular, glassy andesite. Phenocrysts (fresh plagioclase up to 1 mm length, and orthopyroxene), are set in a matrix of hydrated glass and hydrothermal siderite and calcite. Third, there are rhyolite lavas, mostly intersected by wells in the centre of the Rotokawa Geothermal Field between -180 and -720 masl (Fig. 4). The lavas are typically massive, with only minor spherulitic and localised flow band textures (Fig. 5L). The thicknesses vary between 33 m and 290 m for two individual lavas found in RK1. The lavas are sparsely porphyritic (<2 vol.% phenocrysts), containing plagioclase (up to 1.5 mm), plus minor pyroxene and rare quartz. Oruahineawe Formation rhyolite. Most Rotokawa wells (except for RK27, RK27L2, RK16, RK17, and RK18) have drilled through thick flows of feldspar-phyric rhyolite lava. These lavas are 13

Journal Pre-proof here newly formalised as the Oruahineawe Formation (Fig. 4), with a type section in RK8 from 432 to -354 masl. These rhyolite lavas represent a series of individual and coalesced rhyolitic bodies and here are collectively linked to and named after the surficial Oruahineawe dome in the NE part of the field (RK8, Fig. 2, is located on the flank of the dome). As well as contrasts in hydrothermal alteration, variations in flow-banding, brecciation, spherulitic, and pumiceous textures are apparent, but the lavas are petrographically similar. They are weakly porphyritic (<2 vol %) with phenocrysts of plagioclase, pyroxene, hornblende, and quartz (trace only), plus accessory magnetite, apatite and

of

zircon (Fig. 5M). Xenoliths of altered mafic/intermediate clasts that resemble andesite lava are locally present. A core drilled in RK34 at 500 m intercepts rhyolite which clearly displays a

ro

brecciated, matrix supported texture along its upper contact (Fig. 5N). In this rock, there are

-p

abundant angular fragments of perlitic glass, with replacement by adularia, quartz, and pyrite

re

preferentially along margins interpreted as quenched surfaces. The original nature of the glassy matrix is obscured by secondary hydrothermal alteration. The fine matrix totally fills interstices in

lP

the breccia, and is thus unlike normal subaerial dome breccia carapaces (Fink and Anderson, 2000).

na

We thus suggest that this occurrence represents material that was intruded into and quenched by water saturated material, in part auto-brecciated. Any pre-existing sediment incorporated shows no

Jo ur

compositional or textural contrast with the fragmented rhyolite material. Fulljames rhyolites. Petrographically distinct from all other rhyolites at Rotokawa are named by previous workers as the Fulljames 1 and Fulljames 2 rhyolite lavas (Fig. 4), intersected in the west of the field by wells RK16, 17, 18, and 27. The proposed type section for this formation is in well RK16 with the Fulljames 1 Rhyolite from 150 to 105 masl, and the Fulljames 2 Rhyolite from 50 to -290 masl. The two lavas are separated by a ~50 m thick package of breccia overlying volcaniclastic sediments. The breccia is comprised of devitrified rhyolite lava and obsidian fragments, supported in a coarse ash-grade matrix and may be related to emplacement of the overlying lava. Sediment and tuff layers within the deeper half of the package include carbonaceous fine sandstone. The shallower Fulljames 1 lava contains phenocrysts of pyroxene, hornblende, plagioclase and quartz and is characterised by perlitic textured obsidian and, less commonly, flowbanding and spherulitic textures. The Fulljames 2 lava contains fractured and resorbed phenocrysts 14

Journal Pre-proof of quartz, zoned feldspar, hornblende, hornblende glomerocrysts, orthopyroxene and clinopyroxene (including a dark green clinopyroxene that is absent from the Fulljames 1 lava). The Fulljames 2 lava has some intervals with perlitic texture, but mostly is finely vesicular and only rarely shows weakly developed flow banding (Fig. 5O). Huka Falls Formation. This formation is defined as sedimentary and volcaniclastic rocks occurring between the Waiora and Oruanui formations (Fig. 4, Grindley, 1965; Rosenberg et al., 2009). These materials were deposited in a shallow lake located in the region between present-day

of

Lake Taupō and Waiotapu (Steiner, 1963; Rosenberg et al., 2009). At Rotokawa the Huka Falls Formation consists of a 10 to 150 m thick succession (Fig. 4) of interstratified mudstone, siltstone

ro

and sandstone. These laminated lacustrine sediments are variably carbonaceous and commonly

-p

contain dark grey volcanic lithics. Locally the fine sediments are interbedded with crystal-rich

re

(quartz and plagioclase) poorly consolidated sandstones that contain rare pumice grains. The distinction between upper, middle and lower subunits of the formation at Wairakei (Grindley, 1965;

4.5. Oruanui Formation

na

lP

Rosenberg et al., 2009), is not clear at Rotokawa.

Jo ur

At Rotokawa, the Oruanui Formation (Fig. 4) is mostly represented by non-welded ignimbrite consisting of pumice, lithics (andesite, greywacke, and rhyolite) and crystals (feldspar, quartz, pyroxene, minor hornblende) in a pale brown fine-ash-rich matrix. It has a maximum preserved thickness within the geothermal field of 40 m but, based on occurrences beyond the field, has been substantially eroded in the Rotokawa area during post-eruptive lake break-out flood, fluvial, and aeolian processes (Wilson, 1991; Manville and Wilson, 2004).

4.6. Parariki Breccia The Parariki Breccia (Fig. 4) is the collective name given to young phreatic eruption breccia deposits at Rotokawa, as proposed by Collar and Browne (1985). The breccias are the products of hydrothermal eruptions that postdate the Oruanui Formation and range from ~20,000 to 3,700 years in age (Collar and Browne, 1985; Browne and Lawless, 2001). Rotokawa produced one of the 15

Journal Pre-proof largest known Holocene hydrothermal eruptions in New Zealand 6,060 ± 60 14C years ago (conventional Libby age: Collar, 1985), sourced from a site beneath what is now Lake Rotokawa. The surface deposit associated with this eruption extends over an area of 4 km diameter and is in places 11 m thick. Eighty metres of successive phreatic breccias interbedded with lake sediments and tephras have been encountered in RK2 and RK3 (Collar and Browne, 1985) and the total estimated volume is as great as 107 m3. The hydrothermal breccias are poorly-sorted, matrixsupported, polymictic breccia with abundant fragments of rhyolite lava and common clasts that can

of

be visibly matched to lithologies in the Huka Falls, Oruanui and, possibly, Waiora formations.

ro

4.7. Recent tephra deposits

-p

The Oruanui Formation and phreatic breccias are overlain by thin sequences of fluvial

re

sediments, that in turn are draped by a sequence of late glacial loess, and weakly consolidated tephras (all but one being fall deposits) erupted from Taupō and Okataina volcanoes, separated by

lP

thin palaeosols. Non-welded ignimbrite from the 232 ± 10 CE Taupō eruption fills valleys to locally

na

10-20 metres thickness but was mostly eroded out by a catastrophic flood event when Lake Taupō was partially drained (Manville et al., 1999). The youngest material (<1800 years old) consists of

5. U-Pb age data

Jo ur

volcaniclastic alluvium that makes up the modern Waikato River terraces.

Results of the age determinations are summarised in Table 1, with histograms and probability density function (pdf) curves from Isoplot (Ludwig, 2012) presented in Online Appendix 1. Uncertainties on individual ages in Online Appendix 1 are reported to 2 s.d., and the uncertainties on weighted means are 95% confidence intervals, in both cases as yielded by Isoplot. The age determinations on each sample were considered as follows. For lava and pyroclastic samples where the age spectrum was unimodal, that mode was taken as most closely reflecting the eruption age (with older grains taken to be inherited). When bimodal, the younger mode was taken as the eruption age. In the latter case, confidence in the approach taken is given from the fact that the younger mode or fraction of the analysed grains yields age estimates that are, within uncertainty, the same as 16

Journal Pre-proof independently determined eruption ages (e.g., Milicich et al., 2013a). Lead loss is not considered to have influenced the ages: in contrast, Pb-gain with common-Pb is the major issue for most samples but this appears to be accurately resolved by use of the Stacey and Kramers (1975) 207Pb/206Pb value for the common-Pb correction.

5.1. Tahorakuri Formation A rhyolitic tuff sample (RK5-1409, -1086 masl, metres above sea level; Fig. 4) from the

of

volcaniclastic succession of the Tahorakuri Formation was analysed. Fifty-two spots were analysed, of which two returned Mesozoic ages (consistent with greywacke source rock). Of the remaining 50

ro

analyses, 49 were considered acceptable with <30% of the 206Pb attributable to common Pb. The

-p

additional analysis (30.1) was set aside because of its late Miocene age. If the whole population with

re

<30 % of the 206Pb attributable to common Pb (49 spots) is taken, the weighted mean age is 1.87±0.03 Ma (0 of 49 rejected, MSWD = 1.3, probability = 0.053), which is adopted here as the

lP

preferred age.

na

Another tuff sample (RK8-1413, -979 masl; Fig 3) from the Tahorakuri Formation was analysed. Data from this sample span a wide range of ages, with large uncertainties arising from the

Jo ur

generally low U concentrations in zircon grains (26 of 35 grains contain <200 ppm U). A weighted mean of all grains is 1.82 ± 0.05 Ma (1 of 35 rejected, MSWD = 2.6), but as the youngest ages are linked to higher proportions of common-Pb (Fig. 6), an alternative approach is to cull the youngest 5 ages. Doing this yields an age estimate of 1.84 ± 0.04 Ma (0 of 30 rejected, MSWD = 1.6, probability = 0.028), which is analytically indistinguishable from the first sample. The solitary rhyolite lava within the Tahorakuri Formation in RK8 (sample RK8-1164 at 730 masl) yielded only six analysable zircons, with four grains yielding ages consistent with each other (Fig. 7) and the stratigraphic context (Fig. 4). The weighted mean of these four determinations is 720 ± 90 ka.

5.2. Whakamaru Group

17

Journal Pre-proof Three samples were selected from welded ignimbrite (Fig. 4) considered to be part of the Whakamaru Group. On petrographic grounds, this ignimbrite is inferred to share the 349 ± 4 ka 40

Ar/39Ar age determination from Downs et al. (2014b). The three samples yielded age estimates

from zircons as follows. Thirty-five spots were analysed from sample RK6-1174 (-844 masl), all forming a coherent group with <30% of the 206Pb attributable to common Pb and yielding a weighted mean of 362 ± 17 ka (MSWD = 3.9). For sample RK4-1230 (-878 masl), 29 spots were analysed, of which one yielded an early Pleistocene age, then of the remaining 28, 26 spots had <30% of the 206

of

Pb attributable to common Pb and were accepted. The weighted mean of these ages is 333 ± 13 ka

(1 of 26 rejected, MSWD = 2.4). Sample RK5-1204 (-881 masl) had 30 grains analysed. Of these,

ro

the 25 with <30% of the 206Pb attributable to common Pb yielded a weighted mean of 357 ± 15 ka (1

-p

of 25 rejected, MSWD = 2.3). The age distributions for all three samples are fully consistent with

re

these being Whakamaru Group ignimbrite, with the older mean age from RK4-1230 reflecting some greater proportion of older (but consanguineous) grains, as shown in the zircon age spectrum

5.3. Waiora Formation tuffs

na

lP

presented by Brown and Fletcher (1999).

Jo ur

Two samples were selected from rhyolitic tuffs of the Waiora Formation (Fig. 4), to compare with similar materials documented at Wairakei (Rosenberg et al., 2019) and Ngatamariki (Chambefort et al., 2014). Samples RK5-999 (-676 masl) and RK8-860 (-427 masl) yielded rather complex age spectra, with significant proportions of common Pb. With a cut-off of 50% of the 206Pb attributable to common Pb, sample RK5-999 yielded a weighted mean age of 264 ± 15 ka (2 of 19 rejected, MSWD = 1.9) and RK8-860 a weighted mean age of 233 ± 16 ka (1 of 26 rejected, MSWD = 3.7). Although the latter is just resolvably younger, lowering of the common-Pb cut-off to 28% in sample RK8-860 preferentially eliminates the younger ages to give a mean age of 258 ± 16 Ma (1 of 13 rejected, MSWD = 1.9). On this basis, the ages of the two samples are analytically indistinguishable at c. 260 ka, consistent with their stratigraphic positions (Fig. 4).

5.4. Oruahineawe Formation 18

Journal Pre-proof Three samples were selected from coherent rhyolite lava (Fig. 4) to explore the age range of the latest episode of effusive rhyolitic volcanism in the Rotokawa area (RK5-754 [-431 masl], RK4710 [-360 masl] and RK6-262 [68 masl]). All three presented challenges in dating, due to very low levels of radiogenic (non-common) Pb and the fact that the 230Th correction for disequilibrium formed more than half of most of the age estimates. Data where the 206Pb count levels minus the 2 s.d. errors associated with the count statistics did not exceed zero were set aside and the remaining ages with common Pb proportions of <90 % considered. On this basis, estimates of mean ages of

of

100 ± 18 ka (0 of 6 rejected, MSWD = 0.60), 91 ± 10 ka (2 of 21 rejected, MSWD = 1.6) and 81 ± 14 ka (0 of 9 rejected, MSWD = 1.0) are obtained by Isoplot for samples RK5-754, RK4-710 and

ro

RK6-262, respectively. These ages are the same within error, and suggest this large body of rhyolite

re

-p

(whether one or several units) being emplaced around ~100 ka (Fig. 8).

6.1. Data quality

lP

6. Discussion

na

Although the count statistics from the SIMS analyses are unfavourable for precise age determinations due to the young ages of the often U-poor zircons, the consistency of the inferred

Jo ur

ages with respect to stratigraphic succession gives confidence that the age estimates are accurate. In particular, use of the Stacey and Kramers (1975) value for the common-Pb correction for zircons from geothermally altered rocks in the central TVZ provides the basis for accurate correction of common Pb (Wilson et al., 2010; Milicich et al., 2013a, Chambefort et al., 2014). The appropriateness of this correction reflects the saturation of the rocks in the New Zealand geothermal systems in fluids incorporating leached elements from a large volume of greywacke basement, i.e., a representative crustal value (e.g., Price et al., 2015). An important implication from this, and other studies of TVZ geothermal systems cited in this study, is that detailed U-Pb zircon age dating is the only tool available in building models for ages of geothermal activity (and constrain the sources of the modern geothermal resource) and constructing robust models for utilisation of the geothermal resource. In addition, the third

19

Journal Pre-proof dimension given by drillhole information towards understanding histories of eruption and subsidence at individual fields have built a much more comprehensive view of the history of the central TVZ. We suggest that this approach may prove fruitful in other volcanic/geothermal terranes where rocks hosting igneous zircon can be found. Our experience here, and elsewhere in the TVZ, that U-Pb age dating can be extended in hydrothermally altered rocks to periods as young as 100-200 ka, provided that controls are in place to recognise stratigraphic context (e.g., recognition of intrusive Caxton

of

rhyolites at Kawerau: Milicich et al., 2013a).

ro

6.2. Geological history of the Rotokawa area

Refinement of the volcanic and sedimentary rock stratigraphic succession at Rotokawa

6.2.1. Early andesites

lP

re

geological evolution of the Rotokawa area.

-p

coupled with U-Pb dating of key stratigraphic units provides new constraints on the volcanic and

na

The deeply buried andesitic units have been long recognised for their importance in understanding the early volcanic stratigraphy of the Rotokawa area (e.g., Browne et al., 1992), and

Jo ur

the overall evolution of andesitic volcanism in the central TVZ (Wilson et al., 1995), with two episodes of early andesitic volcanism present in the Rotokawa area. The first, and by far more voluminous (up to 1.6 km thick and present in all deep wells at Rotokawa) matches in age andesite lavas (i.e., older than 1.8-1.9 Ma) present at Ngatamariki and Ohaaki (Fig. 9; Massiot et al., 2011; Eastwood et al., 2013; Chambefort et al., 2014). At Rotokawa and Ngatamariki, this early andesite directly sits on Mesozoic greywacke basement, but at Ohaaki Geothermal Field no andesitegreywacke contact had been drilled. The greywacke basement at Rotokawa is inferred to have had some topographic relief (hundreds of metres, in particular along the Central Field Fault: Fig. 10; Wallis et al., 2013), prior to or during emplacement of the andesite. However, there is relatively little topographic relief apparent in the surface of the andesite (Fig. 9: 100-200 m, once later faulting is accounted for). Although voluminous intermediate rocks erupted prior to the onset of silicic caldera-

20

Journal Pre-proof related volcanism are widely documented worldwide (e.g., Lipman et al., 1970; Wark et al., 1990; Wolff et al., 2005) and there is a widespread perception that a change from andesite-dominated to rhyolite-dominated volcanism reflects a change from 'normal' arc conditions to an extensional regime (e.g., Eichelberger, 1978; Hildreth, 1981; Deering et al., 2011), our observations at Rotokawa imply that this picture is oversimplified for the central TVZ. Firstly, the reconstructed morphology of the deep Rotokawa Andesite does not conform to that of a composite cone, rather, it is consistent with the multiple flows being emplaced as ‘sheet-like’ bodies, creating a lava field with subdued

of

topographic relief. Second, the differential displacement of the upper and lower boundaries of the Rotokawa Andesite implies that normal faulting and rift development accompanied lava eruption

ro

and emplacement into a newly or syn-eruptively created basinal structure. Third, the andesite lavas,

-p

where available in drill core, notably lack breccia and scoriaceous intervals and in this respect

re

greatly contrast with lavas exposed on steep-sided composite cones of similar magnitude such as Ruapehu, located in the southern TVZ (Hackett and Houghton, 1989). At Ngatamariki, drilling

lP

intercepted what is inferred to be the same body of andesite lavas (not formally named by

na

Chambefort et al., 2014) in the south of the field at ~-2000 masl (Fig. 9), but its overall geometry is poorly constrained due to extensive intervals of blind drilling (Eastwood et al., 2013; Chambefort et

Jo ur

al., 2014). The closest well at Rotokawa, RK8, intercepted the Rotokawa Andesite at ~ -1500 masl, with the depth difference being explicable by the effects of subsequent rifting and the general slope of the andesitic pile (Fig. 9).

The second period of early andesitic volcanism in the Rotokawa area is represented by the Nga Awa Purua Andesite that is undated due to lack of zircon, but was erupted sometime after the ~1.8 Ma Tahorakuri Formation ignimbrites. An andesite similarly occurring within the Tahorakuri Formation at the Wairakei Geothermal Field has been dated at 1.65 ± 0.05 Ma (Rosenberg et al., 2019). An amphibole-bearing andesite is also found in drillholes in the southern part of the Ngatamariki geothermal system at a similar stratigraphic level to the Nga Awa Purua Andesite at Rotokawa , and the two lavas may reflect the same eruptive episode. This supports the notion that andesite and rhyolitic volcanism have occurred concurrently in the central TVZ throughout its

21

Journal Pre-proof history (Wilson et al., 1995: TVZ in general; Milicich et al., 2013b: Kawerau; Chambefort et al., 2014: Ngatamariki; Sanders et al., 2013: Wairakei). We note also that two examples of lithic lag breccias accompanying caldera forming events nearby in the central TVZ have a dominance of andesite lithics. These are: to the northeast of Rotokawa, the lag breccia of the Kaingaroa ignimbrite sourced from the Reporoa caldera (Beresford and Cole, 2000) and to the southwest the Oruanui ignimbrite lag breccia from Taupō caldera where exposed east of Lake Taupō (C.J.N. Wilson, unpublished data). Both these eruptions evacuated

of

substantial volumes (in the order of 1-10 km3) of andesite, although the absolute ages of these andesites are not known. Magnetic data from the Reporoa caldera area are interpreted to reflect the

ro

presence of buried andesite on its eastern margin at around -400 masl (Soengkono and Hochstein,

-p

1996), while magnetic and seismic investigations across the eastern margin of the TVZ indicate that

re

this buried andesite does not extend east of the Kaingaroa Fault (Stagpoole, 1994). These observations suggest that any of these and the demonstrably early TVZ andesites terminate against

lP

the eastern margin of the TVZ, implying that there was topographic relief associated with the

na

Kaingaroa Fault from early on in central TVZ history.

Jo ur

6.2.2. Thickness and history implications of the Tahorakuri pile Rocks of the Tahorakuri Formation at Rotokawa are thinner (50-300 m thick) and show less textural and lithological diversity than their equivalents nearby at Ngatamariki Geothermal Field (800 – 1700 m thick: Fig. 9; Chambefort et al., 2014), Wairakei Geothermal Field (> 700 m thick: Rosenberg et al., 2009), or at Waiotapu Geothermal Field (>1000 m thick: Wilson et al., 2010). These thickness variations may be explained if rifting-related subsidence in the Rotokawa area relative to the elevation of the Kaingaroa Plateau, occurred later than areas to the northeast, north and southwest (represented by the aforementioned geothermal fields). These differences cannot simply reflect thinning of Tahorakuri Formation deposits over a high-standing edifice of Rotokawa Andesite, as the available ages of the earliest post-andesite rhyolite tuffs are the same within error at Ngatamariki and Rotokawa (Eastwood et al., 2013; Chambefort et al., 2014).

22

Journal Pre-proof The volcaniclastic rock sequence at Rotokawa is condensed in comparison to that at Ngatamariki, with Tahorakuri Formation rocks dated here at 1.87 ± 0.03 Ma and 1.84 ± 0.04 Ma sitting directly beneath the ~350 ka Whakamaru Group ignimbrite. Only the rhyolite lava (dated here at 720 ± 90 ka) and the undated Nga Awa Purua Andesite formed high-standing features around which the Whakamaru Group ignimbrite accumulated (Fig. 10). In contrast, the Tahorakuri Formation silicic deposits at Ngatamariki include at least 800 m thickness with ages from 700 ka to 900 ka. The period of magmatism represented by the rhyolite dome (and, possibly the andesite) is

of

also represented by the 712 ± 27 ka Rolles Peak andesite 5 km to the east (K-Ar age: Tanaka et al., 1996), the crystal-poor 710 ± 60 ka Waiotapu ignimbrite erupted from the Waiotapu area 40 km to

ro

the north-northeast (40Ar/39Ar age: Houghton et al., 1995), and at Wairakei, a sample from the

-p

Kaiapo rhyolite lava complex dated at 733 ± 25 ka (U-Pb on zircons: Rosenberg et al., 2019). These

re

subaerial deposits are contemporaneous with part of the Ngatamariki intrusive complex and its correlative surficial rhyolite dome (Fig. 9; Chambefort et al., 2014). Like the Nga Awa Purua

na

Group ignimbrite.

lP

Andesite and local ~720 ka rhyolite, the Rolles Peak andesite is surrounded by ~350 ka Whakamaru

At Ngatamariki, greywacke gravels (Waikora Formation) are absent (or possibly not

Jo ur

identified due to poor cuttings conditions), but there are common intervals of resedimented andesite interbedded (Fig. 9) with Tahorakuri Formation rhyolitic material (Chambefort et al., 2014). In contrast, within the Tahorakuri Formation at Rotokawa, there are no resolvable intervals of coarse sediment that contain andesite grains, despite Rotokawa Andesite being at the surface. Possibly, locally sourced andesitic sediments were entirely stripped from the area, but more likely is that there was not sufficient relief for significant local erosion to occur (see section 6.2.1) and sedimentation was dominated by distally-sourced greywacke material (Waikora Formation). This situation contrasts with that the Waikora Formation gravels, which from their coarseness (in geothermal drill core) and lithologic nature, are inferred to represent one or more river systems with headwaters to the east in the North Island’s axial mountain ranges. The formation everywhere predates emplacement the Whakamaru Group ignimbrite and, where more detailed stratigraphic interpretation allows, deposition ceased locally prior to 1 Ma (Kawerau: Milicich et al., 2013a; Ngatamariki: 23

Journal Pre-proof Chambefort et al., 2014). Similar greywacke-derived sediments are now transported NNE by the Rangitaiki River (Fig. 1) following the border faults along the western boundary of the axial ranges (Leonard et al., 2010).

6.2.3. Nature of Whakamaru Group After a local eruptive hiatus of at least 350 kyr (Fig. 4), the Rotokawa area was engulfed by ignimbrite of the Whakamaru Group. This ignimbrite is intersected in most wells; absent only where

of

the ~720 ka Tahorakuri Formation rhyolite and Nga Awa Purua Andesite represent local topographic high features around which the ignimbrite was deposited (Figs. 4 and 10). At Rotokawa, the surface

ro

of the Whakamaru Group ignimbrites appears to be little disturbed by post-emplacement faulting at

-p

Rotokawa. The thickness of the Whakamaru Group at nearby geothermal fields (Ngatamariki,

re

Ohaaki, and Mokai) is comparable to that found at Rotokawa, at around 50 to 500 m, but the top is at deeper levels at Rotokawa, Ohaaki and Mokai (approx. -600 masl) than at Ngatamariki (Fig. 9; -400

lP

masl). The Paeroa Subgroup ignimbrite (339 ± 5 ka; Downs et al., 2014b) present in the Ngatamariki

na

area at around -300 masl, is not distinguishable at Rotokawa (Fig. 9). The Whakamaru Group ignimbrite thins to <100 m thickness to the north at Waiotapu and is at shallow depths (>200 masl)

Jo ur

there. To the southwest of Rotokawa, there is a marked contrast, with the Whakamaru Group ignimbrite in the Wairakei Geothermal Field being up to ~1 km thick and intensively block-faulted (Rosenberg et al., 2009, 2019).

There are two inferences that can be made on the basis of these observations. First, the overall variations in thickness of the Whakamaru Group ignimbrite and lack of syn-eruptive faulting at Rotokawa suggest that the boundary of the Whakamaru caldera as originally mapped in Wilson et al. (1986) is open to re-interpretation. The thickness and faulting offset lead us to consider that the boundary of caldera collapse lies between Rotokawa and Wairakei (as shown in this study, and at Wairakei by Rosenberg et al., 2019). Such a structural feature, orientated roughly north-south would help explain the independent behaviour of fluid flow patterns and compositions between the Wairakei and Rotokawa geothermal fields (O’Sullivan et al., 2009). Second, the greater depths to the top of the Whakamaru Group ignimbrite at Rotokawa, Ohaaki and Mokai fields compared to 24

Journal Pre-proof Ngatamariki and Waiotapu is interpreted to reflect the positions of the fields with respect to the graben structures of the Taupō Fault Belt (Mokai: Seebeck et al., 2014b) and Taupō-Reporoa Basin (Rotokawa, Ohaaki: Downs et al., 2014a) versus an intervening horst (Waiotapu, Ngatamariki: Wilson et al., 2010).

6.2.4. Huka Group Within materials collected together as Huka Group, there are a number of complexities.

of

Grindley (1965) divided the Huka Group into two litho- and chrono-stratigraphic formations, the earlier volcaniclastic dominated Waiora Formation and later lacustrine dominated Huka Falls

ro

Formation. Although this distinction is clear at Wairakei (Grindley, 1965; Rosenberg et al., 2009), at

-p

Rotokawa it is obscure and the five members of the Waiora Formation identified by Grindley (1965)

re

at Wairakei cannot be identified. The Waiora Formation at Rotokawa is also finer grained and thinner than at Wairakei (400 m versus 2500 m, respectively), with a greater relative proportion of

lP

lacustrine material. These differences suggest that Rotokawa is farther from the source(s) of the

na

Waiora pyroclastic materials, and that lacustrine conditions locally persisted through much of this depositional interval. Higher ground was present through this time around the Maroa area, to the

Jo ur

northwest, Ngatamariki to the north, and Reporoa to the northeast, where surficial rhyolite lava domes and associated pyroclastic units have been dated to a period between 296 ± 17 ka and ~150 ka (Fig. 1, Leonard, 2003; Chambefort et al., 2014; Downs et al., 2014a; Downs, 2016). We infer that the resulting lacustrine record at Rotokawa was partially overprinted by an influx of volcaniclastic material from these areas. Local depocenters formed in the Rotokawa area, with one such represented in the southeast of the field by a more widespread siltstone (Fig. 4). Rhyolite domes and basalt lava, inferred to be of similar age, are present within the Waiora Formation at Rotokawa and Ngatamariki, which have now subsided to ~-500 masl at Rotokawa and ~0 masl at Ngatamariki. As with the Waiora Formation, the lacustrine-dominated Huka Falls Formation at Rotokawa is less clearly expressed by contrasting lithology than it is at Wairakei or Ngatamariki. The lithofacies evident at Wairakei and Tauhara (Catell et al., 2016) are not resolvable with current drillhole data at either Rotokawa or Ngatamariki. The thicker deposits of the fine-grained lacustrine 25

Journal Pre-proof rocks at Ngatamariki compared to Rotokawa (110-400 m and <150 m respectively) suggest that the shallow lake in which the sediments were deposited persisted for longer at Ngatamariki, or alternatively that there was a deeper sub-basin present. At Rotokawa, there would also have been an influx of sediments delivered by streams coming into the area from the escarpment edge of the Kaingaroa Plateau, and these are inferred to be represented by fine-grained sandstone interbedded with carbonaceous siltstone. Overall, we suggest that the contrast between the Waiora and Huka Falls formations be regarded as a broad lithological distinction with a change upwards from

of

volcaniclastic-dominated to lacustrine-dominated deposition, respectively, and that the contact between them is not isochronous.

ro

The coherent (if imprecise) ages and petrographic characteristics of the subsurface material

-p

labelled Oruahineawe rhyolite closely link them to each other and to the subaerial dome of

re

Oruahineawe. However, the shallow subsurface rhyolite shows upper contact textures that suggest an intrusive origin, with matrix supported autobreccia and hyaloclastite. The subsurface rhyolite

lP

body is thus interpreted here to be a shallow intrusion rather than a buried, downfaulted subaerial

na

equivalent to the surficial Oruahineawe dome. The elevation differences between the subaerial dome and subsurface materials is hundreds of metres (Fig. 10) and there is demonstrably no evidence for

Jo ur

pre-existing topographic relief or fault displacement of this magnitude between the two rhyolite bodies. The intrusive material shows no significant upwards displacement of the country rocks (cf. Showa-shinzan: Yokoyama, 2002; Hipaua dome at Tauhara: Rosenberg and Kilgour, 2003; Caxton Formation rhyolite at Kawerau: Milicich et al., 2013b) and accommodation is inferred to have taken place through compensatory down-sagging as the magma was delivered to shallower levels.

6.3. Duration of the Rotokawa geothermal system As with the other high-temperature geothermal systems in the central TVZ, there arises the question as to the longevity of the current Rotokawa geothermal system. Magmatic activity can be inferred to have been present beneath the Rotokawa area between 700 and 750 ka from the presence of the solitary rhyolite dome, plus other volcanism nearby. There is then little evidence for local heat sources until eruption of the Oruahineawe subaerial and intrusive rhyolites around 100 ka, when a 26

Journal Pre-proof substantive magma system (i.e., large enough to generate several cubic kilometres of crystal-poor rhyolite) underlay the area. However, there is no clear distinction that can be made between any geothermal system associated with that rhyolite versus the modern system, unlike at Kawerau (Milicich et al., 2018), Ngatamariki (Chambefort et al., 2014; 2017), or Wairakei-Tauhara (Rosenberg et al., 2019). The post-25 ka record at Rotokawa includes multiple interbedded hydrothermal eruption breccias (within the Parariki Breccia; Collar and Browne, 1985). Early hydrothermal eruption

of

breccias correspond in age with linked periods of intensive pyroclastic eruptive activity at Taupō and Tongariro volcanoes (Kohn and Topping, 1978). After a quiet period, the largest breccia corresponds

ro

(6,060 ± 60 14C years B.P.; ~6850 cal. years B.P.: D.J. Lowe, pers. comm.) with re-initiation of

-p

volcanic activity at Taupō (unit F: ~6,150 14C years B.P.; Wilson, 1993). Although there is no nearby

re

magmatism, or evidence of surface volcanism at Rotokawa, this suggests that there was a substantial

6.4. Implications for TVZ evolution

lP

disruption to the system at this age, perhaps linked to Taupō through regional rifting processes.

na

The Rotokawa Geothermal Field is now at low elevations within the Taupō-Reporoa basin,

Jo ur

but from drillhole information coupled with the new age data presented here it is clear that for much of its history, this area was somewhat elevated relative to areas to the north (e.g., Ngatamariki). This is supported by differences in the thickness of the Tahorakuri Formation and the ~1.5 Ma time span represented by the disconformity between the Tahorakuri Formation and Whakamaru Group ignimbrite. The Rotokawa Andesite in the northwest part of the field was buried by old ignimbrites (around 1.85 Ma), but the area saw no significant deposition of material between then and the emplacement of the Whakamaru Group ignimbrite at 350 ka, when the localised 720 ± 90 ka rhyolite lava dome was still exposed at the surface. Subsequent deformation of the top Whakamaru Group ignimbrite surface at other geothermal areas indicates that development of the modern arrangement of two arc-parallel zones of subsidence (Taupō-Reporoa basin; Taupō Fault Belt) separated by a horst dates from about 350 ka. However, reconstruction of faulting rates using the Whakamaru Group surface as a marker indicates that despite the significant amounts of younger subsidence, this 27

Journal Pre-proof has been accomplished by movement along faults not intersected by the Rotokawa Geothermal Field drillholes. Information from the new age dating and geological reconstructions of the Rotokawa area serves to highlight aspects of the broader evolution of the central TVZ. Although there is evidence for south-eastward migration of the arc front across the North Island over the last ~16 Myr (Stern et al., 2006; Stern and Benson, 2011; Seebeck et al., 2014a), the stratigraphy and minimum age of the deep andesites at Rotokawa demonstrate that the present-day arc front moved into its modern

of

position around or prior to 2 Ma and has not moved since. Young migration of the arc thus involved a jump of at least 50 km from its position prior to 2 Ma (Seebeck et al., 2014a) and subsequent

ro

immobility. Averaged rates of arc migration thus represent a misleading impression of steadiness

-p

that is contradicted by the geological record. In addition, a recent compilation (Wysoczanski et al.,

re

2019) of age data from the Havre Trough, several hundred kilometres offshore from the North Island and along strike from the TVZ, emphasises that the history of the Rotokawa area shares features in

lP

common with that area. The onset of large-scale rifting and spreading in the Havre Trough was

na

interpreted to be young (<2 Myr) and was simultaneous across wide areas rather representing than a systematic eastwards younging. There is thus an apparent similarity with the history in the Rotokawa

Jo ur

area presented here, with the jump of the TVZ arc front into its present position prior to 1.8-1.9 Ma (and lack of subsequent migration) and the continuation of rifting and magmatism over a broader area.Large-scale (hundreds of metres of vertical displacement at the base but not the top of the andesite pile) rift faulting commenced prior to and/or accompanied eruption of the early andesites. The distinction drawn by some authors between an arc setting for early andesites versus a rifting setting for later rhyolites is thus not soundly based. Although the voluminous early andesites may represent activity accompanying thermal 'conditioning' of the crust prior to development of largescale silicic magmatic systems (e.g., Price et al., 2005; Wolff et al., 2005), their extrusion accompanied rifting, not pre-dated it.

7. Conclusions

28

Journal Pre-proof A study of the Rotokawa Geothermal Field using zircon U-Pb age determinations, combined with stratigraphic interpretations and petrography has been used to reconstruct the geological history of the Rotokawa area. The age data obtained in this study demonstrate the value of absolute age data in constraining histories of magmatism, volcanism, and faulting, to reconstruct the geological framework within which the geothermal resource can be utilised. Ages as young as c. 100 ka (albeit imprecise) have been obtained that are consistent with the proven stratigraphic context, pushing the younger ages limits of the dating technique.

of

A thick sequence of silicic volcanic products is present at Rotokawa and includes ignimbrites of the Tahorakuri Formation that are among the oldest dated silicic volcanic deposits in

ro

the TVZ. These ignimbrites overlie the Rotokawa Andesite, which is likely to extend at least to the

-p

nearby Ngatamariki Geothermal Field. Syn-depositional faulting of the Rotokawa Andesite suggests

re

that rifting in this part of the TVZ initiated prior to 1.89 Ma, which is the maximum age of dated ignimbrites in the overlying Tahorakuri Formation.

lP

Following burial of the area at 350 ka by ignimbrites of the Whakamaru Group, ignimbrites

na

and sedimentary rocks of the Waiora Formation were deposited over a 150 kyr period, coeval with dome-building activity still represented at the surface in the Maroa dome complex northwest of

Jo ur

Rotokawa but buried at the nearby Ngatamariki and Wairakei geothermal fields. Extensive rhyolitic lava bodies from the ~100 ka (combined ages of 3 samples) Oruahineawe Formation show stratigraphic and petrographic relationships implying both extrusive dome and shallow intrusive emplacement. Dominantly lacustrine sedimentary rocks of the Huka Falls Formation and pyroclastic deposits of the 25.4 ± 0.2 ka Oruanui eruption then cap and seal the system. Locally erupted rhyolite lavas provide evidence of periods where underlying magmatic activity could have provided a local heat source for hydrothermal activity, the earliest of these being present between 700 and 750 ka. In recent times, the more voluminous ~100 ka Oruahineawe Formation rhyolite is a more likely source for heat for a geothermal system that may have been continuously active until the present. However, the heat source of the modern geothermal system remains enigmatic because the youngest dated products of a local magma system are likely too old to still be supplying significant heat (e.g., Cathles, 1977). The post-Oruanui record includes multiple 29

Journal Pre-proof Rotokawa-derived hydrothermal eruption breccia deposits collectively labelled the Parariki Breccia, indicating that vigorous hydrothermal activity has occurred since at least ~25 ka. The timing of the individual units that make up the Parariki Breccia suggest that there may be a tectonically mediated link between volcanic activity at Taupō and triggering of major hydrothermal eruptions at Rotokawa.

Acknowledgements

of

The authors acknowledge funding support from GNS Sciences Strategic Investment Fund “New Zealand’s Geothermal Future”. Mercury NZ Limited and Rotokawa Joint Venture are thanked

ro

for access to rock samples and field data. We thank Peter Holden and Peter Lanc for their expertise

-p

and support in obtaining ion probe age data at the SHRIMP facility of the Australian National

re

University. We also thank Michael Rosenberg and two anonymous reviewers for their useful

lP

thoughts on the manuscript.

References

na

Adams, C.J., Mortimer, N., Campbell, H.J., Griffin, W.J., 2009. Age and isotopic characterisation of Torlesse Supergroup and Waipapa Group metasediments in the central North Island, New Zealand. New Zealand Journal of Geology and Geophysics 52, 149–170.

Jo ur

Addison, S.J., Winick, J.A., Mountain, B.W., Siega, F.L., 2015. Rotokawa reservoir tracer test history. Proceedings, 37th New Zealand Geothermal Workshop, Taupo, New Zealand. Beresford, S.W., Cole, J.W., 2000. Kaingaroa Ignimbrite, Taupo Volcanic Zone, New Zealand: evidence for asymmetric caldera subsidence of the Reporoa Caldera. New Zealand Journal of Geology and Geophysics 43, 471–481. Bibby, H.M., Caldwell, T.G., Davey, F.J., Webb, T.H., 1995. Geophysical evidence on the structure of the Taupo Volcanic Zone and its hydrothermal circulation. Journal of Volcanology and Geothermal Research 68, 29–58. Bloomberg, S., Werner, C., Rissmann, C., Mazot, A., Horton, T., Gravley, D., Kennedy, B., Oze, C., 2014. Soil CO2 emissions as a proxy for heat and mass flow assessment, Taupō Volcanic Zone, New Zealand. Geochemistry, Geophysics, Geosystems 15, 4885–4904. Brotheridge, J.M.A., 1995. Surface manifestations – past and present – of the Rotokawa Geothermal Field, Taupo Volcanic Zone, New Zealand. M.Sc. thesis, University of Auckland, New Zealand, 154 p. Brown, S.J.A., Fletcher, I.R., 1999. SHRIMP U-Pb dating of the preeruption growth history of 30

Journal Pre-proof zircons from the 340 ka Whakamaru Ignimbrite, New Zealand: evidence for >250 k.y. magma residence times. Geology 27, 1035–1038. Browne, P.R.L., Graham, I.J., Parker, R.J., Wood, C.P., 1992. Subsurface andesite lavas and plutonic rocks in the Rotokawa and Ngatamariki geothermal systems, Taupo Volcanic Zone, New Zealand. Journal of Volcanology and Geothermal Research 51, 199–215. Browne, P.R.L., Lawless, J.V., 2001. Characteristics of hydrothermal eruptions, with examples from New Zealand and elsewhere. Earth-Science Reviews 52, 299-331. Cathles, L., 1977. An analysis of the cooling of intrusives by ground-water convection which includes boiling. Economic Geology 72, 804–826. Cattell, H.J., Cole, J.W., Oze, C., 2016. Volcanic and sedimentary facies of the Huka Group arc-

of

basin sequence, Wairakei-Tauhara geothermal field, New Zealand. New Zealand Journal of Geology and Geophysics 59, 236-256.

ro

Chambefort, I., Bignall, G. (editors), 2016. Taupo Volcanic Zone geothermal systems, New Zealand:

-p

exploration, science and development. Geothermics 59, 147–356. Chambefort, I., McCoy-West, A., Ramirez, L.E., Rae, A.J., Bignall, G., 2011. Evidence for

re

magmatic fluid pulses into the Rotokawa geothermal system. Proceedings, 33th New Zealand Geothermal Workshop, Auckland, New Zealand.

lP

Chambefort, I., Lewis, B., Wilson, C.J.N., Rae, A.J., Coutts, C., Bignall, G., Ireland, T.R., 2014. Stratigraphy and structure of the Ngatamariki geothermal system from new zircon U-Pb

na

geochronology: implications for Taupo Volcanic Zone evolution. Journal of Volcanology and Geothermal Research 274, 51–70.

Chambefort, I., Lewis, B., Simpson, M.P., Bignall, G., Rae, A.J., Ganefianto, N., 2017. Ngatamariki

Jo ur

geothermal system: magmatic to epithermal transition in the Taupo Volcanic Zone, New Zealand. Economic Geology 112, 319–346. Christiansen, R.L., 2001. The Quaternary and Pliocene Yellowstone Plateau volcanic field of Wyoming, Idaho, and Montana. US Geological Survey Professional Paper 729-G. Cole, B., Legmann, H., 1998. The Rotokawa Geothermal Project: a high pressure, sustainable and environmentally benign power plant. Geothermal Resources Council Transactions 22, 509– 513. Cole, J.W., 1990. Structural control and origin of volcanism in the Taupo Volcanic Zone, New Zealand. Bulletin of Volcanology 52, 445–459. Cole, J.W., Lewis, K.B., 1981. Evolution of the Taupo-Hikurangi subduction system. Tectonophysics 72, 1-21. Collar, R.J., 1985. Hydrothermal eruptions in the Rotokawa geothermal system, Taupo Volcanic Zone, New Zealand. Geothermal Institute, University of Auckland, New Zealand, Report GIR 014, 97 p.

31

Journal Pre-proof Collar, R.J., Browne, P.R.L., 1985. Hydrothermal eruptions at the Rotokawa geothermal field, Taupo Volcanic Zone, New Zealand. Proceedings, 7th New Zealand Geothermal Workshop, University of Auckland, New Zealand, 171–175. Conway, C.E., Leonard, G.S., Townsend, D.B., Calvert, A.T., Wilson, C.J.N., Gamble, J.A., Eaves, S.R., 2016. A high-resolution 40Ar/39Ar lava chronology and edifice construction history for Ruapehu volcano, New Zealand. Journal of Volcanology and Geothermal Research 327, 152– 179. Conway, C.E., Gamble, J.A., Wilson, C.J.N., Leonard, G.S., Townsend, D.B., Calvert, A.T., 2018. New petrological, geochemical and geochronological perspectives on andesite-dacite magma genesis at Ruapehu volcano, New Zealand. American Mineralogist 103, 565–581.

of

Dalrymple, G.B., Grove, M., Lovera, O.M., Harrison, T.M., Hulen, J.B., Lanphere, M.A., 1999. Age and thermal history of the Geysers plutonic complex (felsite unit), Geysers geothermal field,

ro

California: a 40Ar/39Ar and U–Pb study. Earth and Planetary Science Letters 173, 285–298.

-p

Deering, C.D., Bachmann, O., Dufek, J., Gravley, D.M., 2011. Rift-related transition from andesite to rhyolite volcanism in the Taupo Volcanic Zone (New Zealand) controlled by crystal-melt

re

dynamics in mush zones with variable mineral assemblages. Journal of Petrology 52, 2243– 2263.

lP

Downs, D.T., 2016. Mihi Breccia: a stack of lacustrine sediments and subaqueous pyroclastic flows within the Taupo Volcanic Zone, New Zealand. Journal of Volcanology and Geothermal

na

Research 327, 180-191.

Downs, D.T., Rowland, J.V., Wilson, C.J.N., Rosenberg, M.D., Leonard, G.S., Calvert, A.T., 2014a. Evolution of the intra-arc Taupo-Reporoa Basin within the Taupo Volcanic Zone of New

Jo ur

Zealand. Geosphere 10, 185–206.

Downs, D.T., Wilson, C.J.N., Cole, J.W., Rowland, J.V., Calvert, A.T., Leonard, G.S., Keall, J.M., 2014b. Age and eruptive center of the Paeroa Subgroup ignimbrites (Whakamaru Group) within the Taupo Volcanic Zone of New Zealand. Geological Society of America Bulletin 126, 1131–1144.

DSIR, 1985. Sheet U17 — Wairakei. Electrical resistivity map of New Zealand 1:50,000. Nominal Schlumberger array spacing 500 m. Geophysics Division, Department of Scientific and Industrial Research, Wellington New Zealand. Eastwood, A.A., Gravley, D.M., Wilson, C.J.N., Chambefort, I., Oze, C., Cole, J.W., Ireland, T.R., 2013. U-Pb dating of subsurface pyroclastic deposits (Tahorakuri Formation) at Ngatamariki and Rotokawa geothermal fields. Proceedings, 35th New Zealand Geothermal Workshop, Rotorua, New Zealand. Eichelberger, J.C., 1978. Andesitic volcanism and crustal evolution. Nature 275, 21–27. Fink, J.H., Anderson, S.W., 2000. Lava domes and coulees. Encyclopedia of Volcanoes, Academic Press, San Diego, pp. 307-319. 32

Journal Pre-proof Giggenbach, W.F., 1995. Variations in the chemical and isotopic compositions of fluids discharged from the Taupo Volcanic Zone, New Zealand. Journal of Volcanology and Geothermal Research 68, 89–116. Gravley, D.M., Wilson, C.J.N., Rosenberg, M.D., Leonard, G.S., 2006. The nature and age of Ohakuri Formation and Ohakuri Group rocks in surface exposures and geothermal drillhole sequences in the central Taupo Volcanic Zone, New Zealand. New Zealand Journal of Geology and Geophysics 49, 305–308. Grindley, G.W., 1965. The geology, structure, and exploitation of the Wairakei geothermal field, Taupo, New Zealand. New Zealand Geological Survey Bulletin 75, 1-131. Grindley G.W., Browne P.R.L., 1968. The subsurface geology of the Broadlands geothermal field.

of

New Zealand Geological Survey Report 34, 1-41.

Hackett, W.R., Houghton, B.F., 1989. A facies model for a Quaternary andesitic composite volcano:

ro

Ruapehu, New Zealand. Bulletin of Volcanology 51, 51–68.

-p

Hedenquist, J.W., Mroczek, E.K., Giggenbach, W.F., 1988. Geochemistry of the Rotokawa geothermal system: Summary of data, interpretation and appraisal for energy development.

re

N.Z. DSIR Chemistry Division, technical note 88/6, 63p. Hildreth, W., 1981. Gradients in silicic magma chambers: implications for lithospheric magmatism.

lP

Journal of Geophysical Research 86, 10153–10192. Hochstein, M.P., 1995. Crustal heat transfer in the Taupo Volcanic Zone (New Zealand): comparison

na

with other volcanic arcs and explanatory heat source models. Journal of Volcanology and Geothermal Research 68, 117-151.

Hogg, A.G., Lowe, D.J., Palmer, J., Boswijk, G., Bronk Ramsey, C., 2012. Revised calendar date for

Jo ur

the Taupo eruption derived by 14C wiggle-matching using a New Zealand kauri 14C calibration data set. The Holocene 22, 439–449. Houghton, B.F., Wilson, C.J.N., McWilliams, M.O., Lanphere, M.A., Weaver, S.D., Briggs, R.M., Pringle, M.S., 1995. Chronology and dynamics of a large silicic magmatic system: central Taupo Volcanic Zone, New Zealand. Geology 23, 13–16. Hurwitz, S., Lowenstern, J.B., 2014. Dynamics of the Yellowstone hydrothermal system. Reviews of Geophysics 51, 375–411. Kissling, W.M, Weir, G.J., 2005. The spatial distribution of the geothermal fields in the Taupo Volcanic Zone, New Zealand. Journal of Volcanology and Geothermal Research 145, 136– 150. Kohn, B.P., Topping, W.W., 1978. Time-space relationships between late Quaternary rhyolitic and andesitic volcanism in the southern Taupo Volcanic Zone, New Zealand. Geological Society of America Bulletin 89, 1265–1271. Krupp, R.E., Seward, T.M., 1987. The Rotokawa geothermal system, New Zealand; an active epithermal gold-depositing environment. Economic Geology 82, 1109–1129. 33

Journal Pre-proof Krupp, R., Browne, P.R.L. and Henley, R.W., 1986. Rotokawa Geothermal Field. Monograph Series on Mineral Deposits, 26, 47-55. Stuttgart, Germany: Gebruder Borntraeger. Langridge, R.M., Ries, W.F., Litchfield, N.J., Villamor, P., Van Dissen, R.J., Barrell, D.J.A., Rattenbury, M.S., Heron, D.W., Haubrock, S., Townsend, D.B. and Lee, J.M., 2016. The New Zealand active faults database. New Zealand Journal of Geology and Geophysics 59, 86-96. Leonard, G.S., 2003. The evolution of Maroa Volcanic Centre, Taupo Volcanic Zone, New Zealand. Ph.D. thesis, University of Canterbury, Christchurch, New Zealand. Leonard, G.S., Begg, J.G., Wilson, C.J.N. (compilers), 2010. Geology of the Rotorua area: scale 1:250,000. Institute of Geological & Nuclear Sciences 1:250,000 geological map 5. Institute of Geological & Nuclear Sciences Limited, Lower Hutt, New Zealand.

of

Lipman, P.W., Steven, T.A., Mehnert, H.H., 1970. Volcanic history of the San Juan Mountains, Colorado, as indicated by potassium-argon dating. Geological Society of America Bulletin 81,

ro

2329–2352.

-p

Ludwig, K., 2009. SQUID 2: A User’s Manual, revision of 12 April, 2009. Berkeley Geochronology Center Special Publication 5.

re

Ludwig, K.R., 2012. User's Manual for Isoplot 3.75-4.15: A Geochronological Toolkit for Microsoft Excel. Berkeley Geochronology Center Spec. Pub. 4 (revision January 30, 2012), 75 pp.

lP

Manville, V., Wilson, C.J.N., 2004. The 26.5 ka Oruanui eruption, New Zealand: a review of the roles of volcanism and climate in the post-eruptive sedimentary response. New Zealand

na

Journal of Geology and Geophysics 47, 525–547. Manville, V., White, J.D.L., Houghton, B.F., Wilson, C.J.N., 1999. Paleohydrology and sedimentology of a post-1.8 ka breakout flood from intracaldera Lake Taupo, North Island,

Jo ur

New Zealand. Geological Society of America Bulletin 111, 1435–1447. Massiot, C., Bignall, G., Alcaraz, S., van Moerkerk, H., Sepulveda, F., Rae, A., 2011. The history of the Ohaaki Geothermal Field – in 3D. Proceedings, New Zealand Geothermal Workshop 2011, Auckland, New Zealand.

McNamara, D.D., Sewell, S., Buscarlet, E. and Wallis, I.C., 2016. A review of the Rotokawa geothermal field, New Zealand. Geothermics 59, 281–293. Milicich, S.D., Hunt, T., 2007. Thermal features of the Rotokawa Geothermal Field. GNS Science Consultancy Report, 2007/90. Milicich, S.D., Wilson, C.J.N., Bignall, G., Pezaro, B., Charlier, B.L.A., Wooden, J.L., Ireland, T.R., 2013a. U-Pb dating of zircon in hydrothermally altered rocks of the Kawerau Geothermal Field, Taupo Volcanic Zone, New Zealand. Journal of Volcanology and Geothermal Research 253, 97–113. Milicich, S.D., Wilson, C.J.N., Bignall, G., Pezaro, B., Bardsley, C., 2013b. Reconstructing the geological and structural history of an active geothermal field: a case study from New Zealand. Journal of Volcanology and Geothermal Research 262, 7–24. 34

Journal Pre-proof Milicich, S.D., Chambefort, I., Wilson, C.J.N., Charlier, B.L.A., Tepley, F.J., 2018. The hydrothermal evolution of the Kawerau geothermal system, New Zealand. Journal of Volcanology and Geothermal Research, 353, 114–131. Mortimer, N., 1995. Origin of the Torlesse Terrane and coeval rocks, North Island, New Zealand. International Geology Review 36, 891–910. O'Sullivan, M.J., Yeh, A., Mannington, W.I., 2009. A history of numerical modelling of the Wairakei geothermal field. Geothermics 38, 155-168.Price, L., Powell, T.S., Atkinson, L., 2011. Geothermal fluid evolution at Rotokawa: hydrothermal alteration indicators. Geothermal Resources Council Transactions 35, 977–982. Price, R.C., Gamble, J.A., Smith, I.E.M., Stewart, R.B., Eggins, S.M., Wright, I.C., 2005. An

of

integrated model for the temporal evolution of andesites and rhyolites and crustal development in New Zealand's North Island. Journal of Volcanology and Geothermal Research 140, 1–24.

ro

Price, R.C., Gamble, J.A., Smith, I.E.M., Maas, R., Waight, T., Stewart, R.B., Woodhead, J., 2012.

-p

The anatomy of an andesite volcano: a time–stratigraphic study of andesite petrogenesis and crustal evolution at Ruapehu volcano, New Zealand. Journal of Petrology 53, 2139–2189.

re

Price, R.C., Mortimer, N., Smith, I.E.M., Maas, R., 2015. Whole-rock geochemical reference data for Torlesse and Waipapa terranes, North Island, New Zealand. New Zealand Journal of

lP

Geology and Geophysics 58, 213-228.

Ring, U., Mortimer, N., Deckert, H., 2019. Critical-wedge theory and the Mesozoic accretionary

na

wedge of New Zealand. Journal of Structural Geology 122, 1-10. Risk, G.F., 2000. Electrical resistivity of the Rotokawa Geothermal Field, New Zealand. Proceedings, 22nd New Zealand Geothermal Workshop, 121-126.

Jo ur

Rosenberg, M.D., Kilgour, G.N. 2003. Explosive volcanism at Tauhara Volcanic Complex: implications for the emplacement of Hipaua Dome. Abstracts, Geological Society of New Zealand 2003 Annual Conference. Geological Society of New Zealand Miscellaneous Publication 116A, 167.

Rosenberg, M.D., Bignall, G., Rae, A.J., 2009. The geological framework of the Wairakei–Tauhara geothermal system, New Zealand. Geothermics 38, 72–84. Rosenberg, M.D., Wilson, C.J.N., Bignall, G., Ireland, T.R., Sepulveda, F., Charlier, B.L.A., 2019. Structure and evolution of the Wairakei-Tauhara geothermal system (Taupo Volcanic Zone, New Zealand) revisited with a new zircon geochronology. Journal of Volcanology and Geothermal Research (under review). Rowland, J.V., Wilson, C.J.N., Gravley, D.M., 2010. Spatial and temporal variations in magmaassisted rifting, Taupo Volcanic Zone, New Zealand. Journal of Volcanology and Geothermal Research 190, 89–108.

35

Journal Pre-proof Sanders, F., Chambefort, I., Rosenberg, M., Bignall, G., Rae, A., Sepulveda, F., 2013. Characterisation and stratigraphic correlation of the andesites encountered in the Wairakei Geothermal Field. Proceedings, 35th New Zealand Geothermal Workshop, Rotorua, 6 pp. Schmitt, A.K., Grove, M., Harrison, T.M., Lovera, O.M., Hulen, J.B., Walters, M., 2003. The Geysers–Cobb Mountain magma system, California (part 1): U–Pb zircon ages of volcanic rocks, conditions of zircon crystallization and magma residence times. Geochimica et Cosmochimica Acta 67, 3423–3442. Seebeck, H., Nicol, A., Giba, M., Pettinga, J., Walsh, J., 2014a. Geometry of the subducting Pacific plate since 20 Ma, Hikurangi margin, New Zealand. Journal of the Geological Society, London 171, 131–143.

Taupo Rift, New Zealand. Tectonics 33, 1178–1199.

of

Seebeck, H., Nicol, A., Villamor, P., Ristau, J., Pettinga, J., 2014b. Structure and kinematics of the

ro

Sewell, S.M., Addison, S.J., Hernandez, D., Azwar, L., Barnes, M.L., 2015. Rotokawa conceptual

-p

model update 5 years after commissioning of the 138 MWe NAP Plant. Proceedings, New Zealand Geothermal Workshop, Taupo, New Zealand, paper 098.

re

Schinteie, R., Campbell, K.A., Browne, P.R., 2007. Microfacies of stromatolitic sinter from acidsulphate-chloride springs at Parariki Stream, Rotokawa geothermal field, New Zealand.

lP

Palaeontologia Electronica 10, 1-33.Soengkono, S., Hochstein, M.P., 1996. Interpretation of magnetic anomalies over the Reporoa Geothermal Field, Taupo Volcanic Zone, New Zealand.

na

Proceedings 18th New Zealand Geothermal Workshop. Stacey, J.S., Kramers, J.D., 1975. Approximation of terrestrial lead isotope evolution by a two-stage model. Earth and Planetary Science Letters 26, 207–221.

Jo ur

Stagpoole, V.M., 1994. Interpretation of refraction seismic and gravity data across the eastern margin of the Taupo Volcanic Zone, New Zealand. Geothermics 23, 500–510. Steiner, A., 1963. The rocks penetrated by drillholes in the Waiotapu thermal area, and their hydrothermal alteration. New Zealand Department of Scientific and Industrial Research Bulletin 155, 26–34.

Stern, T., Benson, A., 2011. Wide-angle seismic imaging beneath an andesitic arc: Central North Island, New Zealand. Journal of Geophysical Research 116, B09308. Stern, T.A., Stratford, W.R., Salmon, M.L., 2006. Subduction evolution and mantle dynamics at a continental margin: central North Island, New Zealand. Reviews of Geophysics 44, RG4002. Tanaka, H., Turner, G.M., Houghton, B.F., Tachibana, T., Kono, M., McWilliams, M.O., 1996. Palaeomagnetism and chronology of the central Taupo Volcanic Zone, New Zealand. Geophysical Journal International 124, 919–934. Vandergoes, M.J., Hogg, A.G., Lowe, D.J., Newnham, R.M., Denton, G.H., Southon, J., Barrell, D.J., Wilson, C.J.N., McGlone, M.S., Allan, A.S.R., Almond, P.C., 2013. A revised age for the Kawakawa/Oruanui tephra, a key marker for the Last Glacial Maximum in New Zealand. 36

Journal Pre-proof Quaternary Science Reviews 74, 195–201. Wallis, I.C., Bardsley, C.J., Powell, T.W., Rowland, J.V., O’Brien, J.M., 2013. A structural model for the Rotokawa Geothermal Field, New Zealand. Proceedings 35th New Zealand Geothermal Workshop. Rotorua, New Zealand. Wark, D.A., Kempter, K.A., McDowell, F.W., 1990. Evolution of waning subduction related magmatism, northern Sierra Madre Occidental, Mexico. Geological Society of America Bulletin 102, 1555–1564. Wilson, C.J.N., 1991. Ignimbrite morphology and the effects of erosion: a New Zealand case study. Bulletin of Volcanology 53, 635–644. Wilson, C.J.N., 1993. Stratigraphy, chronology, styles and dynamics of late Quaternary eruptions

of

from Taupo volcano, New Zealand. Philosophical Transactions of the Royal Society of London A343, 205–306.

ro

Wilson, C.J.N., Rowland, J.V., 2016. The volcanic, magmatic and tectonic setting of the Taupo

-p

Volcanic Zone, New Zealand, reviewed from a geothermal perspective. Geothermics 59, 167– 186.

re

Wilson, C.J.N., Houghton, B.F., Lloyd, E.F., 1986. Volcanic history and evolution of the MaroaTaupo area, central North Island. In: I.E.M. Smith (ed), Late Cenozoic Volcanism in New

lP

Zealand. Royal Society of New Zealand Bulletin 23, 194–223. Wilson, C.J.N., Houghton, B.F., McWilliams, M.O., Lanphere, M.A., Weaver, S.D., Briggs, R.M.,

na

1995. Volcanic and structural evolution of Taupo Volcanic Zone, New Zealand: a review. Journal of Volcanology and Geothermal Research 68, 1–28. Wilson, C.J.N., Charlier, B.L.A., Fagan, C.J., Spinks, K.D., Gravley, D.M., Simmons, S.F., Browne,

Jo ur

P.R.L., 2008. U–Pb dating of zircon in hydrothermally altered rocks as a correlation tool: application to the Mangakino geothermal field, New Zealand. Journal of Volcanology and Geothermal Research 176, 191–198. Wilson, C.J.N., Gravley, D.M., Leonard, G.S., Rowland, J.V., 2009. Volcanism in the central Taupo Volcanic Zone, New Zealand: tempo, styles and controls. In: Studies in Volcanology: The Legacy of George Walker. Special Publications of IAVCEI 2, 225–247. Wilson, C.J.N., Charlier, B.L.A., Rowland, J.V., Browne, P.R.L., 2010. U–Pb dating of zircon in subsurface, hydrothermally altered pyroclastic deposits and implications for subsidence in a magmatically active rift: Taupo Volcanic Zone, New Zealand. Journal of Volcanology and Geothermal Research 191, 69–78. Winick, J., Powell, T., Mroczek, E., 2011. The natural-state geochemistry of the Rotokawa reservoir. Proceedings, 41st New Zealand Geothermal Workshop, Auckland, New Zealand. Wolff, J.A., Rowe, M.C., Teasdale, R., Gardner, J.N., Ramos, F.C., Heikoop, C.E., 2005. Petrogenesis of pre-caldera mafic lavas, Jemez Mountains volcanic field (New Mexico, USA). Journal of Petrology 46, 407-439. 37

Journal Pre-proof Wysoczanski, R., Leonard, G., Gill, J., Wright, I., Calvert, A., McIntosh, W., Jicha, B., Gamble, J., Timm, C., Handler, M., Drewes-Todd, E., 2019. Ar-Ar age constraints on the timing of Havre Trough opening and magmatism. New Zealand Journal of Geology and Geophysics 62, 1-7. Yokoyama, I, 2002. Growth mechanism of the 1944 lava dome of Usu volcano in Hokkaido, Japan. Proceedings of the Japan Academy, Series B 78, 6-11. Zeh, A., Ovtcharova, M., Wilson, A.H., Schaltegger, U., 2015. The Bushveld Complex was emplaced and cooled in less than one million years – results of zirconology, and geotectonic

Jo ur

na

lP

re

-p

ro

of

implications. Earth and Planetary Science Letters 418, 103–114.

38

Journal Pre-proof Table captions Table 1. List of the samples used for U-Pb dating and summary of age parameters.

Figure captions Figure 1: Simplified map of the Taupō Volcanic Zone (TVZ) showing the locations of the geothermal systems as defined by anomalously low electrical resistivity (<25 Ωm: Bibby et al., 1995), WK: Wairakei, NM: Ngatamariki, OH: Ohaaki, MK: Mokai, WT: Waiotapu. The boundary of the young (350 ka-present) TVZ and caldera locations are after Wilson et al. (1986, 2009). Active faults are from the GNS active fault database (Langridge et al., 2016). The blue box indicates the inset detail for the cross section location in Figure 9.

of

Figure 2: Detail of the Rotokawa area, with well heads (prefixed by RK in text), simplified surface geology from Leonard et al. (2010) and the 30 Ωm resistivity boundary zone of Risk (2000).

ro

Active faults are from the GNS active fault database (Langridge et al., 2016). Blue lines labelled

-p

A-A’ and B-B’ are for cross sections shown in Figure 4.

Figure 3: Comparative stratigraphic columns detailing earlier stratigraphy by previous authors

re

(summarised by McNamara et al., 2016 and references therein) and the revised stratigraphy from this study.

lines shown in Figure 2.

lP

Figure 4: Cross sections of the geological formations of the Rotokawa geothermal system, along

na

Figure 5: Representative macro- and microscopic photographs of key Rotokawa stratigraphic units. Abbreviations: Px; pyroxene, Amg; amygdale; Pl; plagioclase, Amp; amphibole, Qz; quartz, Py; pyrite, Ol; olivine. (A, B) Torlesse greywacke (A) metasandstone litharenite and (B) argillite. (C)

Jo ur

Porphyritic Rotokawa Andesite with plagioclase and pyroxene partially replaced respectively by albite, illite, calcite with some epidote, and adularia, and chlorite. The groundmass consists of fine secondary quartz and chlorite. (D) Hyaloclastite andesitic breccia matrix detail showing primary hematite replaced by secondary pyrite. (E) Cross polarised microscopic photograph of Nga Awa Purua Andesite (NAP) cuttings showing a porphyritic texture with characteristic amphibole and plagioclase phenocrysts. Amphibole is partially replaced by chlorite, calcite and iron oxide, plagioclase is partially replaced by albite, quartz and illite. The groundmass consists dominantly of secondary quartz and chlorite. (F) Fiamme and plagioclase-bearing silicic tuff of the Tahorakuri Formation. The unit is strongly altered to illite, chlorite, quartz, calcite with rare pyrite. (G) Polymictic clast-supported poorly-sorted Waikora Formation conglomerate. Rounded clasts are dominantly greywacke fragments in an illite-rich clay-size matrix. (H) Crystal-rich Whakamaru Group ignimbrite. Abundant quartz and plagioclase are relatively preserved from the alteration. (I) Crystal-poor plagioclase-, quartz-bearing silicic tuff altered to quartz, albite, adularia, illite and rare pyrite. (J) Backscattered secondary electron photograph of an unaltered basalt scoria fragment. The texture is porphyritic to trachytic with abundant unaltered 39

Journal Pre-proof plagioclase, and rare olivine and pyroxene phenocrysts in glass. Circular vesicles are also abundant. (K) SEM image of a basalt clast enveloped in tuffaceous Waiora Formation sediment. Diatoms adhering to the basalt glass support the interpretation that the basalt was emplaced into a lake. (L) Cross polarised microscopic photograph of flow-banded plagioclase-bearing rhyolite (Waiora Formation rhyolite). The devitrified groundmass consists mainly of oxides, quartz, and feldspar. (M) Pyroxene-, plagioclase-bearing rhyolite (Oruahineawe Formation rhyolite). Rare pyroxene is replaced by clay and oxide. Plagioclase is relatively well preserved. (N) Adulariacemented monomictic syn-volcanic clast-supported hyaloclastite breccia (Oruahineawe Formation rhyolite). Clasts consist of perlitic flow-banded rhyolite fragments. Small angular quenched glass fragments are partially altered in the cement. (O) Pumiceous cuttings representing

of

the carapace of the Fulljames 2 Rhyolite dome. (P) Thin section image of polymictic breccia with clasts of variably altered rhyolite, tuff and sedimentary rock of the Parariki Breccia.

ro

Figure 6: Plot of the 230Th-corrected 206Pb/238U ages for sample RK8-1413 from the Tahorakuri

-p

Formation. The negative correlation between the age returned and the proportions of common Pb suggest that the 5 youngest ages can be culled, to yield an estimate of 1.84 ± 0.05 Ma for this

re

sample. See Online Appendix 1 for data

Figure 7: Weighted mean plot for the four consistent age determinations from RK8-1164 used to

lP

derive a best-estimate age 720 ± 90 Ma. See Online Appendix 1 for data. Figure 8: Combined cumulative weighted mean plot for analysed zircons for the three samples from

na

the Oruahineawe Formation rhyolite: samples RK5-754, RK4-710 and RK6-262. An overall age estimate of ~100 ka is adopted here. See Online Appendix 1 for data. Figure 9: Cross section through the Ngatamariki and Rotokawa geothermal fields. The Nga Awa

Jo ur

Purua Andesite is not shown here as it lies off the line of the cross section. Wells are projected onto the cross section. The inset map shows the location of cross section line, with the area covered indicated in Figure 1. The resistivity boundaries are from DSIR (1985) for the Ngatamariki Geothermal Field and Risk (2000) for the Rotokawa Geothermal Field. Ages are from this study, Eastwood et al., (2013), and Chambefort et al. (2014). Figure 10: Time-sequence landscape restoration of the Rotokawa area since the emplacement of the Rotokawa Andesite.

Supplementary data Supplementary data to this article can be found online: Online appendix 1: U-Pb data and complete results. Online appendix 2: Rotokawa Geothermal Field well logs stratigraphy.

40

Journal Pre-proof Table 1 Samples analysed for U-Pb dating of zircons. Depth

95% confidence

Number of analyses

Sample name

Well

mRF

masl

Lithology

Formation

Age

interval, ±

MSWD

total

ages used

RK4-710

RK4

710

-360

Coherent rhyolite

Oruahineawe Formation rhyolite

91 ka

10

1.6

36

19

RK4-1230

RK4

1230

-878

Tuff

Whakamaru Group ignimbrite

333 ka

13

2.4

29

25

RK5-754

RK5

754

-431

Coherent rhyolite

Oruahineawe Formation rhyolite

100 ka

18

0.60

17

6

RK5- 999

RK5

999

-676

Tuff

Waiora Formation

264 ka

15

1.9

28

17

RK5-1204

RK5

1204

-887

Tuff

Whakamaru Group ignimbrite

357 ka

2.3

30

25

RK5-1409

RK5

1409

-1086

Tuff

Tahorakuri Formation

1.87 Ma

1.3

52

49

RK6-1174

RK6

262

68

Coherent rhyolite

Oruahineawe Formation rhyolite

81 ka

14

1.0

26

9

RK8-860

RK8

860

-427

Tuff

Whakamaru Group ignimbrite

362 ka

17

3.9

35

33

RK8-1164*

RK8

1164

-730

Coherent rhyolite

Phenocryst-poor rhyolite lava

e

0.030

720 ka

90

0.26

6

4

RK8-1413

RK8

1413

-979

Tuff

Tahorakuri Formation

1.84 Ma

0.04

1.6

35

30

1612

-1275

Tuff

1.89 Ma

0.02

0.82

36

18

l a

Eastwood et al. (2013) RK06-01

RK6

n r u

Tahorakuri Formation

r P

o r p

f o

15

See Figure 2 for well locations. Depths are in metres below the well head (mRF), while depths masl are relative to sea level. In the lithology, ‘Coherent rhyolite’ is used informally to refer to quartz-bearing material that was coherent (i.e., lava or intrusive material) rather than a welded ignimbrite or other pyroclastic rock (‘Tuff’). The total number of analyses is all those obtained; ‘viable’ analyses refer to those Quaternary grains with acceptable levels of common-Pb (see text), excluding Mesozoic greywacke grains (see Electronic Appendix 1). * Indicative age only from 4 ages with coherent values.

o J

41

Journal Pre-proof

Declaration of interests ☒ The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Jo ur

na

lP

re

-p

ro

of

☐The authors declare the following financial interests/personal relationships which may be considered as potential competing interests:

42

Journal Pre-proof Highlights 

Detailed U-Pb zircon chronology for the Rotokawa geothermal system.



Voluminous early andesite volcanism followed by rhyolitic volcanism with considerable temporal overlap.



Rifting and subsidence in the TVZ began prior to 1.9 Ma.



Subsidence, punctuated and episodic, with most occurring during andesite

In the young record, temporal links between hydrothermal eruptions at Rotokawa

na

lP

re

-p

ro

and volcanism at Taupō.

Jo ur



of

emplacement and postdating 350 ka.

43

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5r1

Figure 5r2

Figure 6

Figure 7

Figure 8