https://doi.org/10.1130/G46250.1

Manuscript received 7 November 2018 Revised manuscript received 11 May 2019 Manuscript accepted 15 May 2019

© 2019 The Authors. Gold Open Access: This paper is published under the terms of the CC-BY license. Published online 10 June 2019

Recycling of depleted continental mantle by subduction and plumes at the Hikurangi Plateau large igneous province, southwestern Pacific Ocean K. Mochizuki1, R. Sutherland2, S. Henrys3, D. Bassett3, H. Van Avendonk4, R. Arai5, S. Kodaira5, G. Fujie5, Y. Yamamoto5, N. Bangs4 and D. Barker3 1Earthquake Research Institute (ERI), University of Tokyo, Tokyo 113-0032, Japan 2SGEES Victoria University of Wellington, PO Box 600, Wellington 6140, New Zealand 3GNS Science, PO Box 30368, Lower Hutt 5040, New Zealand 4Institute for Geophysics, University of Texas at Austin, Austin, Texas 78758-4445, USA 5Japan Agency for Marine-Earth Science and Technology (JAMSTEC), Yokohama 236-0001, Japan

ABSTRACT 2013). The initial data set was acquired in 2009– Seismic reflection and refraction data from Hikurangi Plateau (southwestern Pacific Ocean) 2010 aboard M/V Reflect Resolution with a 98 L require a crustal thickness of 10 ± 1 km, seismic velocity of 7.25 ± 0.35 km/s at the base of the airgun source. Multichannel reflection data were crust, and mantle velocity of 8.30 ± 0.25 km/s just beneath the Moho. Published models of grav- acquired as part of the PEG09 survey (SAHKE-1) ity data that assume normal crust and mantle density predict 5–10-km-thicker crust than we with 10 km streamer length. The SAHKE line was observe, suggesting that the mantle beneath Hikurangi Plateau has anomalously low density, then re-acquired with 100 m shot spacing and with which is inconsistent with previous suggestions of eclogite to explain observations of high seismic 16 ocean-bottom seismometers (OBSs) deployed velocity. The combination of high seismic velocity and low density requires the mantle to be at 5 km spacing across the trench slope and un­ highly depleted and not serpentinized. We propose that Hikurangi Plateau formed by decom- deformed part of the Hikurangi Plateau. A further pression melting of buoyant mantle that was removed from a craton root by subduction, held 10 OBS records were acquired in 2017 aboard beneath 660 km by viscous coupling to slabs, and then rose as a plume from the lower mantle. R/V Marcus G Langseth using a 108 L airgun Ancient Re-Os ages from mantle xenoliths in nearby South Island, New Zealand, support this source, as part of the Seismogenesis at Hikurangi hypothesis. Erosion of buoyant depleted mantle from craton roots by subduction and then re- Integrated Research Experiment (SHIRE). cycling in plumes to make new lithosphere may be an important global geochemical process. OBS data were processed using a minimum- phase Butterworth frequency filter (ramp fre­ INTRODUCTION Coffin, 2004) have all been invoked. We present quencies 2–4 and 8–20 Hz) and spatial trace Hikurangi, Manihiki, and Ontong Java Pla­ new geophysical data from Hikurangi Plateau, the amplitude balancing. In the wide-angle data, teaus (southwestern Pacific Ocean) were em­ southernmost part of this large igneous province we recognized three different phases that we placed coevally and possibly contiguously at ca. (LIP), where we are able to resolve crustal and interpret as (1) refractions through sedimentary 120 Ma (Mahoney et al., 1993; Coffin and Eld­ upper-mantle structure, and hence provide new (Psed) and crustal layers (Pg), (2) refractions holm, 1993). If they formed together, they con­ light on the origin of the plateau (Fig. 1). through the mantle (Pn) of Hikurangi Plateau, stitute the largest igneous province preserved Hikurangi Plateau is actively subducting be­ and (3) a reflected phase (PmP) from the base at Earth’s surface (Davy et al., 2008; Taylor, neath North Island, New Zealand, and can be of Hikurangi Plateau crust (Moho). The collec­ 2006). Chemical and petrological signatures of recognized at depth as a geophysical anomaly tive data set comprises 26 receiver-gathers, from basalts from the plateaus show similarities and on seismic tomography images (Reyners et al., which ~34,000 travel-time interpretations were differences, indicating a compositionally hetero­ 2011). Seismic reflection images (Fig. 2) and made. The picking uncertainties associated with geneous plume source (Fitton and Godard, 2004; consideration of the stratigraphic and magmatic each of these phases were visually estimated to Golowin et al., 2018; Hoernle et al., 2010; Mor­ history of New Zealand reveal that it was pre­ be 40–80 ms (Psed and Pg), 100 ms (Pn) and timer and Parkinson, 1996; Timm et al., 2011). viously subducted beneath the Chatham Rise, 120–150 ms (PmP) respectively. The “Greater Ontong Java Event” included vol­ before 105–85 Ma, which is when Gondwana We constructed our model of crustal struc­ canism in the Nauru, East Mariana, Lyra, and margin subduction ceased in this region (Bland ture by first performing first-arrival travel-time possibly northwest Central Pacific Basins, and et al., 2015; Bradshaw, 1989; Davy et al., 2008; tomog­raphy (Fujie et al., 2006) along the west­ thus covered ~1% of Earth’s surface area (Coffin Sutherland and Hollis, 2001). ern portion of the transect (0–80 km) where and Eldholm, 1994; Mahoney et al., 1993), but OBSs were closely spaced. We matched the there is no consensus on what caused the event: GEOPHYSICAL DATA AND MODEL depth section of the model to the two-way time mantle plumes (Mahoney et al., 1993), mid-ocean We collected wide-angle seismic data along of the reflection section, so that we assigned a ridges, eclogite recycled by mantle circulation a 260-km-long transect during the Seismic Array velocity to each layer bounded by a significant (Anderson, 2005), and bolide impacts (Ingle and Hikurangi Experiment (SAHKE) (Henrys et al., interface picked on the seismic reflection sec­

CITATION: Mochizuki, K., et al., 2019, Recycling of depleted continental mantle by subduction and plumes at the Hikurangi Plateau large igneous province, southwestern Pacific Ocean: Geology, v. 47, p. 795–798, https://​doi​.org​/10​.1130​/G46250.1

Geological Society of America | GEOLOGY | Volume 47 | Number 8 | www.gsapubs.org 795

Downloaded from http://pubs.geoscienceworld.org/gsa/geology/article-pdf/47/8/795/4793769/795.pdf by guest on 23 September 2021 East Mariana Basin teau subducted beneath southern North Island (Henrys et al., 2013; Tozer et al., 2017). Lyra Basin

Nauru Basin High velocities of 8.3–8.9 km/s are found at relatively shallow depth (30–60 km) beneath on­ 0 Central Pacific Basin shore New Zealand (Galea, 1992; Haines, 1979; Ontong Java Kayal and Smith, 1984; Reyners et al., 2006). On this basis, it has been proposed that the base of Figure 1. Location of Manihiki Hikurangi Plateau is composed of eclogite, and 10°S RR Ontong Java, Manihiki, and Hikurangi Plateaus; hence the base of the crust may reach as much ­Nauru, East Mariana, Lyra, as 65 km depth beneath central South Island Central Pacific Basins; (Reyners­ et al., 2011), where Hikurangi Plateau and Robbie Ridge (RR). was subducted during the Cretaceous. Our results 20°S Inset shows location of are not consistent with the hypothesis of a high-

Lord Howe Rise eastern part of Seismic Array Hikurangi Experi- Vp eclogite lower crust. We observe moderately South Fiji Basin ment (SAHKE) transect high Vp values, and it may be that Vp is higher at analyzed in this paper. greater depth, but eclogite cannot be the explana­ 30°S Pacific Ocean NZ—New Zealand. tion for two reasons: (1) the density of eclogite (~3.6 g/cm3) is higher than that of normal mantle New Rapuhia 3 Zealand (~3.3 g/cm ), which is inconsistent with gravity Scarp models (Davy et al., 2008); and (2) the model 400SS NZ Hikurangi depth to the top of high-Vp material (20–24 km) SAHKE is shallower than that predicted for the basalt- Chatham Rise 160 180°E 190°E 200°E eclogite phase transition (>1.1–1.5 GPa; >30– 40 km) (Hacker et al., 2003). We interpret the combination of high Vp tion. We extended each interface east along the tered by adjusting the Moho depth and seismic and low density, which is what exists beneath seismic reflection profile (Fig. 2A) and converted velocity at the base of the crust. The crustal veloc­ the part of Hikurangi Plateau that we have sur­ two-way time into depth assuming that similar ity shown in Figure DR2 is the value just above veyed, as mantle that is highly depleted by melt velocities exist along each interface. Using a for­ the Moho. Mantle velocity was defined immedi­ extraction. A similar pattern of buoyant high-Vp ward modeling approach (Zelt and Smith, 1992; ately beneath the Moho, with a velocity gradient mantle is observed beneath continents (Poupinet Zelt, 1999), the layers were adjusted to replicate prescribed by linearly increasing mantle velocity et al., 2003) and understood from xenoliths and first arrivals interpreted from OBSs deployed by 0.1 km/s between the Moho and 30 km depth. laboratory experiments to be depleted perido­ across Pegasus Basin. After each adjustment, Slices through the parameter volume near the op­ tite, i.e., harzburgite or dunite. High degrees our forward model was converted to two-way timal solution are shown in Figure 2D and Figure of depletion beneath ancient cratons is likely time and compared with the multi-channel seis­ DR2. RMS misfit values are normalized by divid­ because the mantle was hotter and degrees of mic (MCS) section to ensure consistency with ing by the minimum value. The 95% confidence melting were greater earlier in Earth history the structural constraints imposed (Fig. DR1 in region (determined by an F-test) shows that the (Carlson et al., 2005; Pollack, 1986). Xenoliths the GSA Data Repository1). Our final step was crustal thickness is 10 ± 1 km, crustal velocity from nearby South Island are predominantly to use reflected phases from the base of the crust is 7.25 ± 0.35 km/s, and mantle velocity just be­ of peridotite rather than eclogite composition and refracted phases through the mantle to de­ neath the Moho is 8.30 ± 0.25 km/s. (McCoy-­West et al., 2013; Scott et al., 2014), velop an optimized model for the thickness of and depleted mantle is inferred beneath the east­ Hikurangi Plateau crust and the seismic velocity HIKURANGI PLATEAU CRUST ern margin of Manihiki Plateau from basalt geo­ of the underlying mantle (Zelt and Smith, 1992; AND MANTLE chemistry (Ingle et al., 2007). Zelt, 1999). The final seismic velocity model The crustal structure of Hikurangi Plateau has fits 30,000 observations with root-mean-square previously been inferred from modeling of grav­ ORIGIN OF HIKURANGI PLATEAU (RMS) misfit of 55 ms (Figs. DR3–DR8). ity and topography, assuming that it is a basaltic We propose that buoyancy associated with The trade-off between crustal velocity, crustal large igneous province underlain by mantle with ancient depleted mantle played a role in driving thickness, and mantle velocity was analyzed by normal density (Davy and Wood, 1994). Initial Cretaceous mantle upwelling, and that adiabatic calculating how RMS misfit varied with adjust­ results suggested that crust of the plateau near decompression caused melt extraction that was ments to parameters. The crust of Hikurangi the Chatham Rise (Fig. 1) was ~15 km thick, and limited by previous depletion to form the ob­ Plateau was modeled as two layers, with a high subsequent work using constraints from high- served thickness of Hikurangi Plateau. velocity gradient (P wave velocity, Vp, of 5.6– quality seismic reflection data indicated that it Hikurangi Plateau crust was subducted be­ 6.5 km/s) in the upper crust overlying a more could be as thick as 20 km (Davy et al., 2008). neath Chatham Rise and southeastern South gradual velocity gradient through the lower crust. The crustal thickness we find of 10 ± 1 km is ­Island during the Cretaceous (Bradshaw, 1989), Crustal thickness and crustal velocity were al­ significantly less than that of these models, even and hence we expect that lower crust and ­mantle if 1–2 km of low-velocity (~4.5 km/s) overly­ in eastern South Island may be similar to what 1GSA Data Repository item 2019287, illustrated ing material (beneath horizon H in Fig. 2) is we observe on our geophysical transect. The old­ model fits to representative ocean-bottom seismom­ included. Hence, combined gravity and topog­ est continental rocks in nearby eastern South eters, is available online at http://​www.geosociety​ .org​ ​ raphy data require the negative mass anomaly, Island are Carboniferous (<360 Ma) (Mor­ /datarepository​/2019/, or on request from editing@​ modeled as crustal root by Davy et al. (2008), timer, 1995, 2004), but Re-Os isotope ages from geosociety​.org. All SAHKE (http://​dx​.doi​.org​/10​ .21420​/S3H9​-PQ92) and SHIRE (http://dx​ ​.doi​.org​ to be located in the mantle. Our geophysical es­ ­mantle xenoliths are 1.6–2.3 Ga (McCoy-West /10​.21420​/TQ67​-8F60) data presented in this study timate of crustal thickness is better constrained, et al., 2013). A province of craton-like litho­ are available upon request. but consistent with estimates for Hikurangi Pla­ spheric mantle is recognized beneath South

796 www.gsapubs.org | Volume 47 | Number 8 | GEOLOGY | Geological Society of America

Downloaded from http://pubs.geoscienceworld.org/gsa/geology/article-pdf/47/8/795/4793769/795.pdf by guest on 23 September 2021 0 Figure 2. A: Interpreted seismic reflection EAST COAST BASIN PEGASUS BASIN CHATHAM RISE 1 section showing active subduction beneath Hikurangi Trough 2 East Coast Basin, and inactive Cretaceous Deformation Hikurangi subduction faults beneath Chatham Rise 3 Unit 1 front Channel Chatham Rise SB M T (southwest Pacific Ocean). E—base of Eo- 4 PP O Hikurangi Unit 4C P Unit 1 5 DS_faults cene; H—top of Hikurangi Plateau (unit 2); margin E 6 Unit 4A/B M—base of Miocene; O—base of Oligocene; e 7 H P—base of Paleocene; PP—base of Pliocene– Active plate interface Unit 3A/B Gondwana plate interfac

Two-way time (s ) Hikurangi Plateau Pleistocene; SB—seabed; T—top of Torlesse 8 A Unit 2 basement (Bland et al., 2015). B: Best-fitting 9 velocity model, with the top of the relatively 0 low-velocity, high-reflectivity volcanic depos- Hikurangi margin 5 its marked as horizon H. Triangles on seafloor (Active acretionary wedge) Chatham Rise show ocean-bottom seismometer (OBS) loca- H tions. Area without sufficient ray coverage is 10 u shaded in gray. Example record sections and Hikurangi Platea ) calculated ray paths for OBSs shown by yel- 15 (Oceanic LIP: 120-100 Ma low triangles are presented in Figures DR3 Depth (km) to DR8 (see footnote 1). LIP—large igneous 20 province; V.E.—vertical exaggeration. C: Ray 25 density plot. D: Trade-off between crustal 23456789 thickness and mantle velocity in inversion re- B V.E. = 2.83 P-wave velocity (km/s) 30 sults. Contours show root-mean-square misfit 0 values normalized by dividing by the mini- C mum value; red star indicates optimal values 5 of the crustal thickness of 10 km and mantle

velocity of 8.30 km/s. 95% confidence region 10 corresponds to normalized root-mean-square 14 D

misfit of 1.6 (F-value 2.6). 15 12 2.8 Depth (km) 20 2 10 2.2 1.6 ­Island (Scott et al., 2014). We propose that it is 25 8 this material that has anomalously high seismic 050100 150200 250300 Crustal thickness (km) Ray density V.E. = 2.83 wave velocities (Reyners et al., 2011). 30 6 050 100 150 8.0 8.4 8.8 How did ancient cratonic mantle get beneath Distance (km) Mantle velocity (km/s) South Island if the crust was only assembled after 360 Ma? Geochemical analyses of mantle xeno­ spreading ridges (Taylor, 2006) and created the We propose that subduction can erode melt- liths from Antarctica, Australia, and Zealandia LIP that includes the Hikurangi Plateau frag­ depleted subcontinental lithospheric mantle, but have been used to suggest that depleted subconti­ ment (Fig. 3). The mantle reservoir apparently that the inherent buoyancy of this material stops nental lithospheric mantle can be eroded from did not mix substantially geochemically or iso­ it from reaching the base of the mantle. Instead, ancient cratons, circulated through the astheno­ topically with surrounding asthenosphere. it accumulates beneath the 660 km phase transi­ sphere, and then returned back to form new sub­ There is evidence from seismic tomogra­ tion, where viscosity increases by a factor of 80– continental lithospheric mantle (Liu et al., 2015). phy and the geoid that breaks in subduction can 100 (Spasojevic et al., 2010a; Steinberger and We suggest that subcontinental cratonic give rise to upwelling mantle plumes rooted at Calderwood, 2006), and viscous coupling to ad­ mantle was eroded from Gondwana lithosphere 660–1200 km depth (Spasojevic et al., 2010b); jacent slab remnants is sufficient to hold it there. by subduction processes during the Phanerozoic. and there is evidence from dynamic topography, The buoyant mantle is released when there is a It was held beneath 660 km depth by viscous the geoid, and seismic tomography that such a break in the supply of subducted slab (Spasojevic coupling to subducted slabs, and then when a Cretaceous system existed beneath southeastern et al., 2010b; Sutherland et al., 2010). We sug­ break in subduction occurred, it rose as a Creta­ South Island, and persists today beneath Antarc­ gest that this was a primary factor involved in the ceous plume that variably impacted one or more tica (Sutherland et al., 2010). formation of Hikurangi Plateau, which involved a plume rooted at 660–1200 km depth. Inter­ Depth LIP Subduction Continental Figure 3. Cartoon illustrating forma- action between the plume and an active ridge (km) Oceanic Oceanic tion of Hikurangi Plateau (southwest produced the thick crust of Ontong Java Plateau Plume rooted at Pacific Ocean) large igneous prov- and thinner crust of Hikurangi Plateau, which 660–1200 km depth ince (LIP); that is, a conventional 500 was further from the spreading center. Erosion plume rooted in the D layer at the Upper mantle ″ of buoyant depleted mantle from craton roots by Lower mantle base of the mantle (Morgan, 1971), versus one rooted at 660–1200 km subduction and then recycling in plumes to make 1000 depth (Spasojevic et al., 2010), where new lithosphere may be an important global geo­ Plume from base of mantle the reservoir of sub-continental chemical process and may create instabilities that mantle (+) is derived from long-lived influence large-scale mantle flow. 1500 subduction beneath the ancient con- tinent. We propose that subduction erodes buoyant depleted mantle from ACKNOWLEDGMENTS 2000 Subcontinental mantle the lithosphere and holds it beneath We appreciate the efforts of the captains and crew of eroded and held down 660 km, where there is a strong vis- the M/V Reflect Resolution, the R/V Marcus Langseth, as subduction continues cous coupling to sinking slabs (-). The and the R/V Tangaroa during the SAHKE and SHIRE plume is released when subduction 2500 cruises. We also thank the technical staff from the D″ dies, and if the plume impacts the Earthquake Research Institute (ERI) at the University spreading ridge, then the Hikurangi of Tokyo, GNS Science (New Zealand), and the Japan Base of mantle 2900 Plateau large LIP is produced. Agency for Marine-Earth Science and Technology

Geological Society of America | GEOLOGY | Volume 47 | Number 8 | www.gsapubs.org 797

Downloaded from http://pubs.geoscienceworld.org/gsa/geology/article-pdf/47/8/795/4793769/795.pdf by guest on 23 September 2021 (JAMSTEC) for their contribution to these projects. Haines, A.J., 1979, Seismic wave velocities in the Letters, v. 212, p. 89–101, https://doi​ .org​ /10​ .1016​ ​ This work was funded by the governments of New uppermost­ mantle beneath New Zealand: New /S0012​-821X​(03)00258​-9​. Zealand, Japan, and the United States (U.S. National Zealand Journal of Geology and Geophysics, v. 22, Reyners, M., Eberhart-Phillips, D., Stuart, G., and Science Foundation grant OCE-1615815 to Van Aven­ p. 245–257, https://​doi​.org​/10​.1080​/00288306​ Nishimura, Y., 2006, Imaging subduction from donk and Bangs), and by individual institutions of .1979​.10424223​. the trench to 300 km depth beneath the central JAMSTEC and ERI. We thank the editor, Chris Clark, Henrys, S., Wech, A., Sutherland, R., Stern, T., ­Savage, North Island, New Zealand, with Vp and Vp/Vs: and anonymous reviewers for constructive feedback. M., Sato, H., Mochizuki, K., Iwasaki, T., Okaya, Geophysical Journal International, v. 165, p. 565– D., and Seward, A., 2013, SAHKE geophysical 583, https://​doi​.org​/10​.1111​/j​.1365​-246X​.2006​ transect reveals crustal and subduction zone .02897​.x​. REFERENCES CITED structure at the southern Hikurangi margin, New Reyners, M., Eberhart-Phillips, D., and Bannister, Anderson, D.L., 2005, Large igneous provinces, delami­ Zealand: Geochemistry Geophysics Geosystems, S., 2011, Tracking repeated subduction of the nation, and fertile mantle: Elements, v. 1, p. 271– v. 14, p. 2063–2083, https://doi​ .org​ /10​ .1002​ /ggge​ ​ ­Hikurangi Plateau beneath New Zealand: Earth 275, https://​doi​.org​/10​.2113​/gselements​.1​.5​.271​. .20136​. and Planetary Science Letters, v. 311, p. 165–171, Bland, K.J., Uruski, C.I., and Isaac, M.J., 2015, Pega­ Hoernle, K., Hauff, F., Van den Bogaard, P., Werner, https://​doi​.org​/10​.1016​/j​.epsl​.2011​.09​.011​. sus Basin, eastern New Zealand: A stratigraphic R., Mortimer, N., Geldmacher, J., Garbe-Schön­ Scott, J.M., Waight, T.E., Van der Meer, Q.H.A., Palin, record of subsidence and subduction, ancient and berg, D., and Davy, B., 2010, Age and geo­ J.M., Cooper, A.F., and Münker, C., 2014, Meta­ modern: New Zealand Journal of Geology and chemistry of volcanic rocks from the Hikurangi somatized ancient lithospheric mantle beneath the Geophysics, v. 58, p. 319–343, https://doi​ .org​ /10​ ​ and Manihiki oceanic Plateaus: Geochimica et young Zealandia microcontinent and its role in .1080​/00288306​.2015​.1076862​. Cosmo­chimica Acta, v. 74, p. 7196–7219, https://​ HIMU‐like intraplate magmatism: Geochemistry Bradshaw, J.D., 1989, Cretaceous geotectonic patterns doi​.org​/10​.1016​/j​.gca​.2010​.09​.030​. Geophysics Geosystems, v. 15, p. 3477–3501, in the New Zealand region: Tectonics, v. 8, p. 803– Ingle, S., and Coffin, M.F., 2004, Impact origin for the https://​doi​.org​/10​.1002​/2014GC005300​. 820, https://​doi​.org​/10​.1029​/TC008i004p00803​. greater Ontong Java Plateau?: Earth and Planetary Spasojevic, S., Gurnis, M., and Sutherland, R., 2010a, Carlson, R.W., Pearson, D.G., and James, D.E., 2005, Science Letters, v. 218, p. 123–134, https://​doi​ Inferring mantle properties with an evolving dy­ Physical, chemical, and chronological charac­ .org​/10​.1016​/S0012​-821X​(03)00629​-0​. namic model of the Antarctica–New Zealand re­ teristics of continental mantle: Reviews of Geo­ Ingle, S., Mahoney, J.J., Sato, H., Coffin, M.F., gion from the Late Cretaceous: Journal of Geo­ physics, v. 43, RG1001, https://​doi​.org​/10​.1029​ Kimura, J.-I., Hirano, N., and Nakanishi, M., physical Research, v. 115, B05402, https://​doi​ /2004RG000156​. 2007, Depleted mantle wedge and sediment .org​/10​.1029​/2009JB006612​. Coffin, M.F., and Eldholm, O., 1993, Scratching the finger­print in unusual basalts from the Manihiki Spasojevic, S., Gurnis, M., and Sutherland, R., 2010b, surface: Estimating dimensions of large igneous Plateau, central Pacific Ocean: Geology, v. 35, Mantle upwellings above slab graveyards linked provinces: Geology, v. 21, p. 515–518, https://​ p. 595–598, https://doi​ .org​ /10​ .1130​ /G23741A​ .1​ .​ to the global geoid lows: Nature Geoscience, v. 3, doi​.org​/10​.1130​/0091​-7613​(1993)021​<0515:​ Kayal, J.R., and Smith, E.G.C., 1984, Upper mantle­ p. 435–438, https://​doi​.org​/10​.1038​/ngeo855​. STSEDO>2​.3​.CO;2​. P-wave velocities in the southeast North Is­ Steinberger, B., and Calderwood, A.R., 2006, Models Coffin, M.F., and Eldholm, O., 1994, Large igne­ land, New Zealand: Tectonophysics, v. 104, of large-scale viscous flow in the Earth’s mantle ous provinces: Crustal structure, dimensions, p. 115–125, https://​doi​.org​/10​.1016​/0040​-1951​ with constraints from mineral physics and surface and external consequences: Reviews of Geo­ (84)90105​-7​. observations: Geophysical Journal International, physics, v. 32, p. 1–36, https://​doi​.org​/10​.1029​ Liu, J., Scott, J.M., Martin, C.E., and Pearson, D.G., v. 167, p. 1461–1481, https://​doi​.org​/10​.1111​/j​ /93RG02508​. 2015, The longevity of Archean mantle residues .1365​-246X​.2006​.03131​.x​. Davy, B., and Wood, R., 1994, Gravity and magnetic in the convecting upper mantle and their role in Sutherland, R., and Hollis, C., 2001, Cretaceous de­ modelling of the Hikurangi Plateau: Marine Geol­ young continent formation: Earth and Planetary mise of the Moa plate and strike-slip motion at the ogy, v. 118, p. 139–151, https://​doi​.org​/10​.1016​ Science Letters, v. 424, p. 109–118, https://​doi​ Gondwana margin: Geology, v. 29, p. 279–282, /0025​-3227​(94)90117​-1​. .org​/10​.1016​/j​.epsl​.2015​.05​.027​. https://​doi​.org​/10​.1130​/0091​-7613​(2001)029​ Davy, B., Hoernle, K., and Werner, R., 2008, Hikurangi Mahoney, J.J., Storey, M., Duncan, R.A., Spencer, <0279:​CDOTMP>2​.0​.CO;2​. Plateau: Crustal structure, rifted formation, and K.J., and Pringle, M., 1993, Geochemistry and Sutherland, R., Spasojevic, S., and Gurnis, M., 2010, Gondwana subduction history: Geochemistry age of the Ontong Java Plateau, in Pringle, M.S., Mantle upwelling after Gondwana subduction Geophysics Geosystems, v. 9, Q07004, https://​ et al., eds., The Mesozoic Pacific: Geology, Tec­ death explains anomalous topography and sub­ doi​.org​/10​.1029​/2007GC001855​. tonics, and Volcanism: American Geophysical sidence histories of eastern New Zealand and Fitton, J.G., and Godard, M., 2004, Origin and evo­ Union Geophysical Monograph 77, p. 233–261. West Antarctica: Geology, v. 38, p. 155–158, lution of magmas on the Ontong Java Plateau, McCoy-West, A.J., Bennett, V.C., Puchtel, I.S., and https://​doi​.org​/10​.1130​/G30613​.1​. in Fitton, J.G., et al., eds., Origin and Evolution Walker, R.J., 2013, Extreme persistence of cra­ Taylor, B., 2006, The single largest oceanic plateau: of the Ontong Java Plateau: Geological Society tonic lithosphere in the southwest Pacific: Paleo­ Ontong Java–Manihiki–Hikurangi: Earth and of London Special Publication 229, p. 151–178, proterozoic Os isotopic signatures in Zealandia: Planetary Science Letters, v. 241, p. 372–380, https://doi​ .org​ /10​ .1144​ /GSL​ .SP​ .2004​ .229​ .01​ .10​ .​ Geology, v. 41, p. 231–234, https://​doi​.org​/10​ https://​doi​.org​/10​.1016​/j​.epsl​.2005​.11​.049​. Fujie, G., Ito, A., Kodaira, S., Tkahashi, N., and .1130​/G33626​.1​. Timm, C., Hoernle, K., Werner, R., Hauff, F., van Kaneda, Y., 2006, Confirming sharp bending of Morgan, W.J., 1971, Convection plumes in the lower den Bogaard, P., Michael, P., Coffin, M.F., and the Pacific plate in the northern Japan trench sub­ mantle: Nature, v. 230, p. 42–43, https://​doi​.org​ Koppers, A., 2011, Age and geochemistry of the duction zone by applying a traveltime mapping /10​.1038​/230042a0​. oceanic Manihiki Plateau, SW Pacific: New evi­ method: Physics of the Earth and Planetary In­ Mortimer, N., 1995, Origin of the Torlesse terrane and dence for a plume origin: Earth and Planetary teriors, v. 157, p. 72–85, https://​doi​.org​/10​.1016​ coeval rocks, North Island, New Zealand: Interna­ Science Letters, v. 304, p. 135–146, https://​doi​ /j​.pepi​.2006​.03​.013​. tional Geology Review, v. 36, p. 891–910, https://​ .org​/10​.1016​/j​.epsl​.2011​.01​.025​. Galea, P., 1992, Observations of very high P-velocities doi​.org​/10​.1080​/00206819409465494​. Tozer, B., Stern, T.A., Lamb, S.L., and Henrys, in the subducted slab, New Zealand, and their re­ Mortimer, N., 2004, New Zealand’s geologi­ S.A., 2017, Crust and upper-mantle structure of lation with the slab geometry: Geophysical Jour­ cal foundations: Gondwana Research, v. 7, ­Wanganui Basin and southern Hikurangi margin, nal International, v. 110, p. 238–250, https://​doi​ p. 261–272, https://doi​ .org​ /10​ .1016​ /S1342​ ​-937X​ North Island, New Zealand as revealed by active .org​/10​.1111​/j​.1365​-246X​.1992​.tb00870​.x​. (05)70324-5. source seismic data: Geophysical Journal Inter­ Golowin, R., Portnyagin, M., Hoernle, K., Hauff, Mortimer, N., and Parkinson, D., 1996, Hikurangi Pla­ national, v. 211, p. 718–740, https://​doi​.org​/10​ F., Werner, R., and Garbe-Schönberg, D., 2018, teau: A Cretaceous large igneous province in the .1093​/gji​/ggx303​. Geochemistry of deep Manihiki Plateau crust: southwest Pacific Ocean: Journal of Geophysical Zelt, C.A., 1999, Modelling strategies and model Implications for compositional diversity of Research, v. 101, p. 687–696, https://​doi​.org​/10​ assessment for wide-angle seismic traveltime large igneous provinces in the Western Pacific .1029​/95JB03037​. data: Geophysical Journal International, v. 139, and their genetic link: Journal of Chemical Geol­ Pollack, H.N., 1986, Cratonization and thermal evolu­ p. 183–204, https://doi​ .org​ /10​ .1046​ /j​ .1365​ ​-246X​ ogy, v. 493, p. 553–566, https://doi​ .org​ /10​ .1016​ /j​ ​ tion of the mantle: Earth and Planetary Science .1999​.00934​.x​. .chemgeo​.2018​.07​.016​. Letters, v. 80, p. 175–182, https://doi​ .org​ /10​ .1016​ ​ Zelt, C.A., and Smith, R.B., 1992, Seismic traveltime in­ Hacker, B.R., Abers, G.A., and Peacock, S.M., 2003, /0012​-821X​(86)90031​-2​. version for 2-D crustal velocity structure: Geophys­ Subduction factory 1. Theoretical mineralogy, Poupinet, G., Arndt, N., and Vacher, P., 2003, Seis­ ical Journal International, v. 108, p. 16–34, https://​

densities, seismic wave speeds, and H2O contents: mic tomography beneath stable tectonic regions doi​.org​/10​.1111​/j​.1365​-246X​.1992​.tb00836​.x​. Journal of Geophysical Research, v. 108, 2029, and the origin and composition of the continental https://​doi​.org​/10​.1029​/2001JB001127​. lithospheric mantle: Earth and Planetary Science Printed in USA

798 www.gsapubs.org | Volume 47 | Number 8 | GEOLOGY | Geological Society of America

Downloaded from http://pubs.geoscienceworld.org/gsa/geology/article-pdf/47/8/795/4793769/795.pdf by guest on 23 September 2021