The inception and progression of melting in a monogenetic eruption: Motukorea Volcano, the Auckland Volcanic Field, New Zealand

Abstract

Export Date: 4 January 2013, Source: Scopus, doi: 10.1016/j.lithos.2012.09.012, Language of Original Document: English, Correspondence Address: McGee, L.E.; School of Environment, The University of Auckland, Private Bag 92019, Auckland, New Zealand; email: lucy.e.mcgee@gmail.com, References: Adam, J., Green, T., Trace element partitioning between mica- and amphibole-bearing garnet lherzolite and hydrous basanitic melt: 1. Experimental results and the investigation of controls on partitioning behaviour (2006) Contributions to Mineralogy and Petrology, 152, pp. 1-17;

Full text

The inception and progression of melting in a monogenetic eruption: Motukorea
Volcano, the Auckland Volcanic Field, New Zealand
Lucy E. McGee
a,
⁎, Marc-Alban Millet
b,1
, Ian E.M. Smith
a
, Károly Németh
c
, Jan M. Lindsay
a
a
School of Environment, The University of Auckland, Private Bag 92019, Auckland, New Zealand
b
School of Geography, Environment and Earth Sciences, Victoria University of Wellington, PO Box 600, Wellington, New Zealand
c
Volcanic Risk Solutions, Massey University, PO Box 11222, Palmerston North, New Zealand
abstractarticle info
Article history:
Received 23 March 2012
Accepted 26 September 2012
Available online 2 October 2012
Keywords:
Auckland Volcanic Field
Monogenetic
Alkalic basalt
Nephelinite
Pb isotopes
Compositional variation through basaltic monogenetic eruptive sequences provides a unique view into the
processes and source heterogeneity of small-scale magmatic systems. A well-exposed, continuous sequence
on Motukorea volcano in the Auckland Volcanic Field, New Zealand, consists of an early tuff ring, scoriaceous
deposits and late lava flows which allow the evolution of the eruption to be studied at very high resolution.
The deposits show a spectrum of basaltic compositions from Mg# 60 nephelinite (early tuff ring) to Mg# 70
alkalic basalt (lava). Within the deposits of each main eruptive phase (i.e. tuff, scoria and lava) very little var-
iation is observed in major element chemistry, suggesting that fractional crystallisation has a limited effect.
Systematic changes in trace element chemistry, however, are significant through the sequence. The major
and trace element features observed through the sequence are inferred to be primarily due to the mixing
of two magma batches, with a two-fold increase in the degree of melting between these. Variation in
Pb-isotopic compositions up-sequence indicates subtle changes in mantle source with samples representing
the start of the eruption displaying higher
207
Pb/
204
Pb than the latter parts of the eruption. This chemical
change coincides with a switch in the mode of eruption, with the arrival at the surface of magmas produced
by larger degrees of partial melting resulting in the beginning of a more effusive eruption phase. The
silica-undersaturated, high total alkali, low Al
2
O
3
and higher
207
Pb/
204
Pb nature of the samples from the
tuff units suggests that these samples were produced by melting of relatively young eclogite domains. The
lower
207
Pb/
204
Pb, higher silica, lower total alkali nature of the samples from the scoria and lava reflects
the exhaustion of these domains and the resultant melting of the surrounding garnet-peridotite matrix.
This detailed study shows that the petrogenesis of small volcanic centres may be far more complex than
their physical volcanology suggests.
© 2012 Elsevier B.V. All rights reserved.
1. Introduction
Monogenetic eruptions of basalt where there has been only limited
modification in magma chambers or by crustal assimilation provide
valuable information regardingthe deeper processes of magma genera-
tion and extraction. The term ‘monogenetic’describes small volume,
typically basaltic volcanoes that have been built in a continuous erup-
tion sequence within a relatively short time span (on the scale of days
up to several decades) (Connor and Conway, 2000; Kienle et al., 1980;
Valentine and Gregg, 2008). An assumption is that monogenetic volca-
noes are built by the eruption of a single compositionally discrete batch
of magma. Recent detailed work on individual centres in volcanic fields,
however, shows that they are more often the product of relatively
complex magmatic processes, commonly involving more than one
magma batch within a single eruptive episode (e.g. Bradshaw and
Smith, 1994; Brenna et al., 2010, 2011; Needham et al., 2011;
Valentine and Gregg, 2008; Valentine and Hirano, 2010; Valentine and
Keating, 2007). Detailed studies of volcanic sequences have investigated
systematic relationships of chemical composition to stratigraphic posi-
tion (and therefore to time in an eruption sequence) and found that
they are far from simple. For example, Strong and Wolff (2003) docu-
mented the compositional changes through several monogenetic se-
quences in the Southern Cascades, USA, and described differences
both within scoria deposits, and between the scoria and lavas of the
same centre. This was attributed to the involvement of several distinct
sources over the course of one eruption. This study (in addition to
others, e.g. Blondes et al., 2008; Brenna et al., 2010; Cebriá et al., 2011;
Garcia et al., 2000; Reiners, 2002; Smith et al., 2008) demonstrates the
significance of monogenetic volcanism in sampling heterogeneous
source regions, and also the complexity of simple sequences.
The observation that chemical compositions, melting processes and
source characteristics can be highly variable from one eruptive phase to
Lithos 155 (2012) 360–374
⁎Corresponding author. Tel.: +64 9 373 7599x88824; fax: +64 9 373 7435.
E-mail addresses: lucy.e.mcgee@gmail.com,l.mcgee@auckland.ac.nz (L.E. McGee),
millet@uchicago.edu (M.-A. Millet), ie.smith@auckland.ac.nz (I.E.M. Smith),
K.Nemeth@massey.ac.nz (K. Németh), j.lindsay@auckland.ac.nz (J.M. Lindsay).
1
Now at: Origins Laboratory, Department of Geophysical Sciences, University of
Chicago and Enrico Fermi Institute, 5734 South Ellis Avenue, Chicago, IL 60637, USA.
0024-4937/$ –see front matter © 2012 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.lithos.2012.09.012
Contents lists available at SciVerse ScienceDirect
Lithos
journal homepage: www.elsevier.com/locate/lithos

another in a single episode shows the importance of detailed sampling
through a complete volcanic sequence. Here we present the results of a
high resolution volcanological and geochemical study of the deposits of
a single monogenetic eruptive centre in the Auckland Volcanic Field
(AVF): Motukorea. The AVF is a Quaternary basaltic intraplate volcanic
field consisting of c. 50 volcanic centres which take the form of scoria
cones, tuff rings and maars. The fieldcoversanareaofc.360km
2
on
the isthmus of Auckland in the North Island of New Zealand (Fig. 1A),
and has been active for c.250 ka (see Lindsay et al., 2011 and references
therein). Individual volcanic centres in the field are of very small vol-
ume and have distinct geochemistry (McGee et al., in review).
Motukorea volcano (Fig. 2) has an exceptionally well-preserved se-
quence of deposits spanning the duration of the eruption with no evi-
dence of hiatuses, and thus can be considered monogenetic in terms
of its physical volcanology. A striking feature of the deposits at
Motukorea is their primitive nature (shown by MgO contents of
9–13 wt.%, Table 1), which allows direct investigation of mantle pro-
cesses. A uranium (U-) series study which focussed on the AVF by
Huang et al. (1997) included some samples from Motukorea which
were used to model a melting scenario for the AVF as a whole, however,
a complete geochemical and isotopic study of the sequence has until
now been lacking. We document compositional variations in major
and trace elements and Pb isotopes through all phases of the eruption,
including the tuff sequence, overlying scoriaceous unit, scoria cone
and lava flows in order to investigate the magmatic conditions that
led to the eruption of a small volume, monogenetic volcanic centre.
2. Geomorphology and physical volcanology of Motukorea
Motukorea is a well-preserved basaltic volcano, the minimum age of
which has been estimated as 7000–9000 years B.P, based on an early
Holocene high stand terracebuilt over lava flows (Bryner, 1991). The is-
land covers an area of just 0.75 km
2
and is c. 900 m wide at its widest
point (Fig. 1B). Its outcrops display a complete volcanic sequence
which is composed of a tuff ring, a central scoria cone (66 m in height,
c. 180 m in diameter), lava flows and several rafted spatter mounds
(referred to here as ‘scoria mounds’)(Figs. 1Band2). Volume estimates
were recently calculated from a Digital SurfaceModel created from spot
height detected by Light Detection and Radar (LiDAR) techniques. The
calculated dense rock equivalent (DRE) volumes for the various volca-
nic units are the following: 0.9×10
6
m
3
for the tuff ring, 2×10
6
m
3
for the scoria cone, and 2.8× 10
6
m
3
for the lava flow including the
rafted spatter/scoria mounds. Correction values used for calculation of
DRE are 30%, 50% and 95% respectively, consistent with previous esti-
mates for the Auckland Volcanic Field (Allen and Smith, 1994; Allen et
al., 1996) (G. Kereszturi, pers. comm; Kereszturi et al., 2012).
We refer to the following eruptive units, in order of eruption: 1) the
basal tuff ring-forming unit (comprising the ‘lower tuff sequence’,
‘mid-scoriaceous unit’and the ‘upper tuff sequence’) outcropping as
10–25 m high cliffs of interbedded tephra layers on the north and
northeast side of the volcanic island, which have been partially eroded
from the northeast at Crater Bay (Figs. 1Band2), 2) the overlying unit
of scoriaceous lapilli beds (‘upper scoriaceous unit’: USU) inferred to
have erupted from the intra-tuff ring scoria cone, 3) the intra-tuff ring
scoria cone edifice including the rafted mounds (‘scoria mounds’), and
4) the lava flows emitted from the central scoria cone region. There
are no soil horizons between eruptive units or other evidence of time
breaks that would suggest that the deposits represent anything other
than a continuous sequence. Because of this we interpret Motukorea
as having formed during a single eruptive episode, and therefore
monogenetic in terms of its physical volcanology (e.g. Németh, 2010;
Valentine and Gregg, 2008).
5km
Rangitoto
channel
Waitemata harbour
Manukau harbour
Auckland
CBD
N
Motukorea
200km N
TVZ
AVF
Volcanic centre
N
100m
?
Lava flows
Scoria
Tuff
Scoriaceous
hills
Scoriaceous
hills
?Log 1
Log 2
A
BCrater
Bay
Scoria cone
Crater
Crater
Hill
Fig. 1. (A) Schematic map of the Auckland Volcanic Field (AVF) showing the location of volcanic centres, including Motukorea. The location of Crater Hill volcano (referred to in
Section 5) is also shown. Inset shows position of the AVF relative to other major volcanic centres in the North Island of New Zealand, including the Taupo volcanic Zone (TVZ).
(B) Geological sketch map of Motukorea Island showing the main volcanic features. Location of logs (Fig. 3) is indicated by dashed boxes.
361L.E. McGee et al. / Lithos 155 (2012) 360–374

2.1. Basal tuff ring sequence and upper scoriaceous unit
The basal tuff ring sequence was logged at two locations: ‘Southeast
Crater Bay’(Log 1) and ‘Northwest Crater Bay’(Log 2) (Figs. 1Band3).
Samples were taken from both logged sections in order to obtain mate-
rial representing the entire tuff sequence (and some units were more
accessible in one than theother). Due to the northeast dip of the tuff de-
posits Log 2 begins at a higher level (B1, correlating with S12 in Log 1),
with the lower section of this unit outcropping below wave base. We
note that Motukorea is thought to have erupted ina non-marine setting
(although possibly estuarine, Allen et al., 1996), therefore the base of
the exposed deposits in Log 1 is unlikely to be the base of the sequence
and information regarding the initial eruptive deposits cannot therefore
be obtained. Several units in the tuff sequence have characteristic fea-
tures whichhelped with correlation between Logs 1 and 2, for example
the large plastically deformed lithics in S14/B2, the juvenile-dominated
lapilli unit S17 (‘mid-scoriaceous unit’: MSU) and shell fossils in S18
(B5) (Fig. 3); however, correlation was more challenging further up-
sequence (S23/B9 upwards) as the characteristics of the two logs dif-
fered. S22 to S29 in Log 1 is dominated by planar beds of juvenile lapilli
interbedded with cross-bedded ash-rich layers, whereas the corre-
sponding units in Log 2 (B6 toB15) are more chaotic and contain abun-
dant large blocks with directional indicators and impact sags. These
observations led to the conclusion that logs 1 and 2 represent laterally
continuous distal and proximal sections, respectively (Fig. 3).
We have divided the basal tuff ring sequence into upper and lower
sections, separated by the distinctive marker bed of scoriaceous ba-
saltic lapilli (S17 –MSU) which ranges from a c. 20 cm thick layer
at Crater Bay to a c. 1.5 m thick bed further to the southeast. In gen-
eral, the tuff sequence comprises fine-grained, cream-beige coloured,
typically cross-bedded layers of weakly to moderately indurated,
poorly to moderately sorted, tuff and lapilli tuff beds, containing
abundant dense juvenile and accidental lithic lapilli. These beds alter-
nate with planar-bedded to massive black-brown, cm to few dm
thick, relatively well-sorted, laterally continuous, juvenile-lapilli-
rich units, an example being the mid-scoriaceous unit, inferred to
represent pyroclastic fall beds (Fig. 3). Interbedding between acci-
dental lithic-rich, dune and cross-bedded tuff and lapilli tuff and
fall-dominated juvenile pyroclast-rich lapilli beds is commonly rhyth-
mic, especially in the upper tuff ring deposit (i.e. above the MSU).
Accidental lithic fragment-rich tuff and lapilli tuff beds and juvenile-
lapilli-rich beds are typically 5–30 cm and 20–100 cm thick, respec-
tively; the thickness of individual layers varies laterally. Within this
stratigraphic framework, we interpret the lower and upper tuff se-
quences as dominantly phreatomagmatic surge deposits.
Juvenile material in the ash-dominated layers of the tuff sequence
ranges from fresh, black, vesicular, sometimes glassy, ragged clasts
(selected for geochemical analysis) to moderately weathered, brown-
ish, non-vesicular, smoother clasts, the latter of which may represent
degassed, recycled juvenile material. Both juvenile clast types range
in size from lapilli to bomb size; the bombs often show plastic defor-
mation and rest in large sags, and many contain small (c. 0.5–2cm)
lithics of sandy-clay material. Petrographically, the juvenile bombs
are glassy, with few olivine phenocrysts, and small inclusions of acci-
dental lithics. Lithic material is abundant throughout the tuff se-
quence; the majority of clasts are a pale sandy to clay material
which sometimes contains relict bedding. These occur as the cores
of bombs, as small inclusions in bombs, or as individual clasts ranging
from c. 1 cm up to large blocks ≤20 cm in diameter. Some of these
blocks are plastically deformed and have impacted surrounding
layers due to water release (Fig. 3). These lithics are thought to be de-
rived from the underlying Waitemata Group sedimentary sequence
(Bryner, 1991), the uppermost country rock in much of the Auckland
area (Edbrooke, 2001). Also present as lithics are larger (c. 10–25 cm)
clasts of orange-coloured sedimentary conglomerate. These are typi-
cally rounded and smooth, but also occur as less coherent clasts.
This type of lithic clast is more abundant in the lower part of the
lower tuff sequence, although it also occurs in the top part of the
upper tuff sequence. These clasts show characteristics such as a
large range in clast colour and type that are typical of the so-called
‘Parnell Volcaniclastic Conglomerate’facies of the Waitemata Group
sedimentary sequence, and are therefore interpreted to have been de-
rived from this unit. A rare clast of potentially ultramafic material
which may be sourced from the Brook Street volcanics terrane –
part of the Dun Mountain Ophiolite Belt which passes beneath
Auckland –was found at the southernmost extent of the tuff se-
quence of the island.
Although the abundance of juvenile basaltic bombs and lapilli in-
creases in the uppermost units of the upper tuff sequence (Fig. 3),
there is a more abrupt change in style in the top unit of the cliff section
(‘upper scoriaceous unit’: USU). This comprises a 4–10 m thick se-
quence of moderately to highly vesicular fine lapilli and coarse blocky
scoria in diffusely massive beds with very rare lithic clasts (Fig. 3). The
deposit does not vary laterally where exposed. Juvenile blocks in this
layer range from 10 cm up to 70 cm, commonly have chilled margins
and are elongate in shape and plastically deformed. Scoria samples are
Fig. 2. Aerial photograph of Motukorea Island taken from the Northeast, showing the well-exposed tuff sequence at Crater Bay, the scoria cone and associated mounds. Photograph
courtesy of Bruce Hayward.
362 L.E. McGee et al. / Lithos 155 (2012) 360–374

Table 1
Representative whole rock major and trace element analyses for rock samples from Motukorea.
Lower tuff sequence MSU Upper tuff sequence
Height (m) 0.5 1 6.25 7.5 9 12 13 21
Sample MBI-1-3 MBI-3-2 MBI-B2-2 MBI-15-2 MBI-17-2 MBI-22-1 MBI-B9-1 MBI-34-1
SiO
2
40.21 41.94 41.15 40.78 39.95 41.21 39.73 41.63
TiO
2
2.83 2.74 2.77 2.78 2.70 2.76 2.80 2.71
Al
2
O
3
11.53 11.60 11.58 11.48 11.18 11.55 11.50 11.73
tot
Fe
2
O
3
15.29 14.91 14.94 15.57 14.76 14.73 14.96 14.55
MnO 0.23 0.22 0.23 0.23 0.22 0.22 0.23 0.22
MgO 10.01 9.54 10.20 9.89 10.57 10.41 10.53 10.50
CaO 11.21 10.88 11.16 10.89 10.75 11.49 11.52 11.21
Na
2
O 4.89 4.58 4.46 4.66 4.61 4.34 4.18 4.60
K
2
O 1.53 1.99 1.87 1.83 1.23 1.55 1.67 1.69
P
2
O
5
1.17 1.11 1.10 1.11 1.05 1.02 1.12 1.01
Total 99.72 99.69 99.67 99.72 99.69 99.69 99.69 99.70
LOI 0.82 0.17 0.22 −0.33 1.61 0.41 1.45 −0.16
Mg# 61 60 62 60 63 62 62 63
Sc 21.7 21.6 22.6 19.7 19.8 23.0 19.7 23.2
V 225 218 236 225 226 238 226 237
Cr 259 247 312 301 314 333 281 340
Ni 188 177 205 183 192 205 188 212
Rb 22.6 37.2 33.4 32.5 11.1 27.5 27.6 30.6
Sr 918 1114 1039 963 1237 1082 971 1019
Y 31.8 31.8 30.8 27.5 25.6 35.0 27.0 28.3
Zr 390 381 370 342 349 349 380 337
Nb 112 110 106 108 98 100 107 99
Cs 0.19 0.57 0.47 0.52 0.89 0.35 0.37 0.57
Ba 313 509 478 386 173 370 406 455
La 84.7 82.2 79.2 78.1 72.8 74.3 76.9 72.7
Ce 146 142 138 148 137 134 144 134
Pr 17.0 16.5 16.0 16.6 15.5 15.2 16.3 14.9
Nd 69.6 67.8 65.8 65.8 61.1 61.1 64.2 60.4
Sm 12.5 12.5 11.8 12.2 11.0 11.3 11.6 11.0
Eu 3.66 3.68 3.52 3.60 3.40 3.36 3.56 3.32
Gd 10.9 10.4 10.3 10.1 9.51 9.68 9.83 9.05
Tb 1.38 1.36 1.36 1.27 1.20 1.21 1.23 1.22
Dy 7.53 7.47 7.22 6.77 6.36 6.50 6.68 6.77
Ho 1.24 1.28 1.20 1.12 1.03 1.13 1.10 1.12
Er 2.88 2.96 2.84 2.67 2.49 2.71 2.63 2.80
Tm 0.36 0.36 0.35 0.32 0.31 0.35 0.33 0.34
Yb 1.97 2.10 1.95 1.69 1.71 1.90 1.75 1.88
Lu 0.26 0.27 0.27 0.22 0.21 0.26 0.22 0.25
Hf 8.16 8.34 7.87 7.28 6.74 7.13 7.07 6.90
Ta 7.84 7.63 7.42 7.14 6.67 6.92 7.10 6.70
Th 11.2 11.0 10.3 10.0 9.19 9.63 9.88 9.41
U 2.40 2.81 2.67 2.68 2.09 2.39 3.10 2.60
206
Pb/
204
Pb 19.328± 0.002 19.297 ± 0.001 19.250 ± 0.001 19.303±0.001 19.249 ± 0.002 19.158 ± 0.001 19.182 ±0.002 19.298 ± 0.001
207
Pb/
204
Pb 15.598± 0.001 15.605 ± 0.001 15.602 ± 0.001 15.602±0.001 15.603 ± 0.002 15.598 ± 0.001 15.600 ±0.001 15.602 ± 0.001
208
Pb/
204
Pb 38.937± 0.003 38.934 ± 0.003 38.886 ± 0.003 38.932±0.003 38.888 ± 0.005 38.800 ± 0.002 38.830 ±0.003 38.923 ± 0.003
Upper scoriaceous unit Scoria mounds Lava
Height (m) 22 23 23 n/a n/a n/a n/a n/a n/a
Sample MBI-SCO-1 MBI-SCO-2 MBI-SCO-4 MBI-SCO-6 MBI-SCO-7 MBI-SCO-9 MBI-Lava-1 MBI-Lava-3 MBI-Lava-4
SiO
2
41.37 42.10 41.89 42.33 42.36 42.44 43.20 42.01 41.89
TiO
2
2.75 2.67 2.66 2.63 2.59 2.60 2.59 2.62 2.63
Al
2
O
3
11.94 12.06 12.00 12.09 12.03 12.19 11.96 11.90 12.07
tot
Fe
2
O
3
14.50 13.97 13.98 13.74 13.47 13.62 13.35 13.61 13.96
MnO 0.22 0.21 0.21 0.20 0.19 0.19 0.19 0.18 0.21
MgO 10.64 11.09 11.17 11.85 12.23 12.32 12.72 13.28 11.63
CaO 11.58 11.48 11.59 11.27 10.94 10.69 11.17 10.86 11.39
Na
2
O 4.50 3.91 3.46 3.45 3.10 2.75 3.15 2.75 3.62
K
2
O 1.37 1.48 1.39 1.07 1.29 1.24 1.24 1.15 1.41
P
2
O
5
0.94 0.80 0.78 0.61 0.61 0.59 0.49 0.50 0.70
Total 99.70 99.69 99.71 99.72 99.73 99.74 99.76 99.77 99.77
LOI −0.11 −0.10 −0.26 0.13 0.61 0.61 −0.29 0.36 −0.14
Mg# 63 65 65 67 68 68 69 70 66
Sc 25.9 28.9 29.5 26.7 27.2 27.4 25.8 26.1 26.6
V 245 250 256 254 258 259 270 270 262
Cr 392 431 485 466 488 493 490 447 452
Ni 215 231 283 263 287 287 304 293 236
Rb 19.4 27.9 28.6 39.5 23.1 20.9 16.9 16.9 23.9
Sr 572 1034 897 669 590 567 529 529 727
Y 20.0 29.7 30.0 30.0 27.0 26.0 19.8 23.4
Zr 318 284 210 235 213 214 200 192 187
Nb 56 86 86 68 60 60 53 53 73
(continued on next page)
363L.E. McGee et al. / Lithos 155 (2012) 360–374

highly vesicular, and contain olivine phenocrysts. Titaniferous (Ti)-
augite is present as a minor groundmass phase.
The characteristics of the basal tuff ring sequence at Motukorea
(Fig. 3) allow some generalisations to be made about the conditions of
the initial phase of the eruption. There is a general transition from
matrix-supported facies at the base to clast-supported facies in the
upper parts of the sequence, corresponding to an increase in clast vesic-
ularity upwards (Allen et al., 1996). This has been interpreted to repre-
sent a trend from a dominantly wet phreatomagmatic fragmentation
eruptive style early in the eruption history of the tuff ring, to a domi-
nantly dry phreatomagmatic to magmatic fragmentation style in the
waning phase of the tuff eruption (e.g. Németh, 2010). This is inferred
to correspond to the gradual exhaustion of the local water supply lead-
ing to an increase in the magma:water ratio (Allen et al., 1996; Bryner,
1991). Similar eruption style changes have been inferred on the basis of
changes in sedimentological features from base to top at other tuff rings
in the AVF (e.g. Crater Hill, Houghton et al., 1999). The relative abun-
dance of thin phreatomagmatic tuff and lapilli tuff layers in the upper
tuff sequence indicates cyclic but localized recharge of water to the sys-
tem to fuel explosive magma-water interaction. The gradual increase of
scoriaceous ash and lapilli layers in the top of the basal tuff ringunit in-
dicate a rapid subsequent drying out of the system allowing explosive
magma fragmentation to dominate. The transition to the purely mag-
matic upper scoriaceous unit indicates further drying out of the system,
either due to the complete exhaustion of water or a significant increase
in the magma:water ratio.
2.2. Scoria cone, rafted scoria mounds and lava flows
The intra-tuff ring scoria cone is an unbreached, circular edifice
c. 70 m high in the northern part of the island. It has a c. 10 m deep cra-
ter, and the rim is higher on the northeast side. There is a small group of
low hills (c. 10–25 m high, Figs.1Band2) to the south and southwest of
the scoria cone composed of scoriaceous agglutinate and lapilli com-
monly displaying lava spatter layering and moderate agglutination
and welding. The implication is that the mounds are rafted sections of
the scoria cone transported on top of lava flows issued from the crater
towards the end of the eruption. The present day circular shape of the
intra-crater scoria coneis inferred to be the result of subsequent healing
of the cone due to penecontemporaneous explosive activity, similar
to those processes described from Red Crater, Arizona (Riggs and
Duffield, 2008).
Lava flows cover the southern part of island (Fig. 1B), and have a
minimum thickness of c. 1.5 m. The lavas have flowed to the south for
c. 1 km; their terminus now lies underwater to the south and south-
west. They are dominantly pahoehoe with rounded vesicles. Polygonal
jointing on the top surface is visible in some outcrops. Lavas are dense
and moderately non-vesicular. Olivine and Ti-augite are present as phe-
nocryst phases –olivine being the larger and most abundant of these.
Plagioclase, Ti-augite, oxides and glass make up the groundmass.
3. Methodology
Thirty-four samples of juvenile material were collected from all key
units of the Motukorea eruption sequence for geochemical analysis.
Twenty-one samples of juvenile material (typically blocks and bombs
from the ash-rich layers and lapilli clasts from the scoriaceous layers)
were collected from the tuff sequence, with samples taken every
c. 0.5–1 m up-sequence, including several from the mid-scoriaceous
unit (Fig. 3). Four samples were taken from the unit overlying the tuff
sequence (USU), five from the scoria mounds, and four samples from
the lava flows. No samples were taken from the scoria cone itself due
to lack of exposed material. When selecting blocks and bombs for geo-
chemistry, an attempt was made to select inclusion-free, glassy, fresh
material, interpreted to represent magma at the time of the eruption.
Seventeen of these samples were analysed for Pb isotopes (Table 1).
To the suite of 34 samples, 10 previously collected and analysed,
well-located samples (from the upper scoriaceous unit, scoria mounds
and lava flows) were included to augment the dataset.
Samples were washed in distilled water and dried at 80 °C before
being crushed in a tungsten carbide mill to b200 μmesh. Major ele-
ments were analysed by XRF on fused glass discs made using Lithium
Borate Spectrachem 12–22 flux, using a Siemens SRS3000 sequential
X-ray spectrometer with a Rh tube at the University of Auckland.
Table 1 (continued)
Upper scoriaceous unit Scoria mounds Lava
Height (m) 22 23 23 n/a n/a n/a n/a n/a n/a
Sample MBI-SCO-1 MBI-SCO-2 MBI-SCO-4 MBI-SCO-6 MBI-SCO-7 MBI-SCO-9 MBI-Lava-1 MBI-Lava-3 MBI-Lava-4
Cs 0.25 0.38 0.39 0.35 0.28 0.28 0.26 0.26 0.33
Ba 273 386 391 342 308 314 234 234 344
La 36.2 63.9 64.3 46.3 39.0 39.1 32.4 32.4 50.9
Ce 70.3 111 112 88.2 75.5 75.6 64.2 64.2 97.8
Pr 8.2 13.0 13.0 10.2 8.7 8.8 7.5 7.5 11.1
Nd 33.1 53.9 54.1 40.6 35.6 35.7 30.8 30.8 43.8
Sm 6.70 10.2 10.1 8.02 7.33 6.96 6.41 6.41 8.32
Eu 2.18 3.10 3.10 2.45 2.29 2.31 2.09 2.09 2.64
Gd 6.15 9.15 9.07 7.07 6.72 6.40 5.89 5.89 7.35
Tb 0.81 1.24 1.26 0.93 0.87 0.86 0.79 0.79 0.99
Dy 4.74 6.86 6.88 5.35 4.85 5.10 4.52 4.52 5.47
Ho 0.80 1.20 1.21 0.92 0.89 0.85 0.80 0.80 0.94
Er 2.11 2.92 2.99 2.34 2.24 2.20 2.05 2.05 2.39
Tm 0.28 0.37 0.40 0.30 0.29 0.29 0.26 0.26 0.30
Yb 1.55 2.13 2.21 1.70 1.64 1.67 1.49 1.49 1.73
Lu 0.20 0.28 0.29 0.23 0.23 0.21 0.20 0.20 0.23
Hf 4.02 6.38 6.44 4.61 4.32 4.27 4.12 4.12 4.83
Ta 3.64 6.01 6.00 4.41 3.96 3.87 3.52 3.52 4.62
Th 4.44 8.31 8.36 5.59 4.75 4.75 4.08 4.08 6.18
U 1.29 2.02 1.95 1.59 1.28 1.31 1.32 1.32 1.84
206
Pb/
204
Pb 19.339±
0.002
19.300±
0.001
19.355±
0.002
19.360±
0.001
19.330±
0.001
19.330±
0.001
19.308±
0.001
19.228±
0.002
19.253±
0.002
207
Pb/
204
Pb 15.604±
0.002
15.604±
0.001
15.595±
0.002
15.591±
0.001
15.589±
0.011
15.595±
0.001
15.590±
0.001
15.593±
0.001
15.594±
0.002
208
Pb/
204
Pb 38.958±
0.004
38.929±
0.003
38.942±
0.005
38.930±
0.002
38.906±
0.003
38.923±
0.003
38.891±
0.003
38.847±
0.003
38.859±
0.004
364 L.E. McGee et al. / Lithos 155 (2012) 360–374

Trace elements were measured on a Laser Ablation Inductively
Coupled Mass Spectrometer (LA-ICP-MS) at the Australian National
University following the procedure of Eggins et al. (1998). XRF discs
of samples were glued with epoxy into stacks of 15, cut, mounted
(30 samples per mount) and polished on the side to be ablated. A
103 μm spot size was used to track down each half of the mount;
NIST 612 was run every 15 samples and used for calibration, and
the silica content obtained by XRF used in data reduction as an inter-
nal standard. BCR-2G was used as a secondary standard. The samples
were run over three sessions (the 10 previously collected samples in
June 2007, the others in December 2010 and June 2011) and each ses-
sion included samples from all eruptive phases; precision in these
BASE
2m
4m
6m
8m
10m
12m
14m
16m
18m
~36m
S1
S2
S3
S4
S5
S6
S7*
S8
S9
S10
S11
S13
S15
S16*
S17
S18
S19
S20
S21
S22
S23
S24
S25
S26
S27
S28
S29
S30
S14
S12
Mid
scoriaceous
unit
Lower
tuff
sequence
Upper
tuff
sequence
Upper
scoriaceous
unit
B1 (S12)
B2
(S13-15)
B3
(S16)
B4 (S17)
B5
(S18-19)
B6
(S20)
B7 (S21)
B8 (S22)
B9 (S23)
B10-13
(S24-27)
B14 (S28)
B15 (S29)
B16 (S30)
S31
S32
S33
S34
BASE
1m
2m
3m
4m
5m
6m
7m
8m
9m
10m
11m
12m
13m
14m
15m
~18m
~26m
Directional
indicators
Lithic sags
and directional
indicators
Bomb sags
Bomb sags
Cross bedded
ash-rich layers
Rhythmic juvenile
and ash beds
Plastic deformation
of lithics, sags
Dune/cross
bedding
Dune/cross
bedding
Log 1: Southeast Crater Bay Log 2: Northwest Crater Bay
SE
Lower tuff
Mid-scoriaceous unit
Upper tuff
Upper scoriaceous unit
Samples taken:
Fig. 3. Simplified logs of the tuff sequence in two locations at Motukorea (see Fig. 1B for locati on of logs) with descriptive features used in determining proximal and distal deposits.
Units in the sequence are labelled ‘S’at Southeast Crater Bay and correlated with units labelled ‘B’in Northwest Crater Bay until S30/B16 is reached then ‘S’labelling system is re-
sumed. Dashed lines indicate correlation of units across logs. Samples were taken from units marked with coloured symbols –these symbols are used in later geochemical plots.
Note change of scale in Log 2 from 9 m. Gap below Upper Scoriaceous Unit in Log 1 is due to the height of the cliff at this point; the unit is correlated with that in Log 2 where the
deposit was more accessible.
365L.E. McGee et al. / Lithos 155 (2012) 360–374

sessions is ≤5%, ≤9% and ≤14% (2SD) respectively, and accuracy is
b7% (b9% for Y, Cs and Tb), b4% (b6% for Sc and Y) and b10% respec-
tively. Precision across all BCR-2G data (n= 64) is b12% (2SD) for all
elements except Y, Lu, Yb and Hf which are b14% (2SD). Accuracy is
≤10% for all elements. XRF data is reported for Cr, V, Ni and Zr. The
Supplementary Data (sheet 2) presents all BCR-2G analyses.
Pb isotopes were prepared and measured at Victoria University of
Wellington (VuW) in ultra-clean lab conditions, using Optima™acids.
Sample powders were leached in hot 6 M HCl for one hour then rinsed
to remove un-bonded Pb after Millet et al. (2008). Powders were
digested in hot concentrated HNO
3
and HF for 24 h, then dried down
and nitrified once. Samples were then taken up in 0.8 M HBr twice,
and centrifuged. Pb was separated in a double pass through a pipette
tip column filled with AG1-X8 resin (see Baker et al. (2004) for proce-
dure). Pb isotopes were measured on a Nu® instruments MultiCollector
ICP-MS at VuW in static mode. Pb isotope measurements used a
sample-standard bracketing method with NBS-981 as the bracketing
standard to correct for instrumental mass-bias and drift (Baker
et al., 2004). JB-2 was run as a secondary standard and measured
as
206
Pb/
204
Pb= 18.3402,
207
Pb/
204
Pb= 15.5621 and
208
Pb/
204
Pb=
38.2755 with 2SD of 166, 239 and 288 ppm respectively (based on
one digestion measured 5 times) close to the reference value of Baker
2
5%
5%
2
4
6
8
10
35 40 45 50
Na2O + K2O
Nephelinite
Basanite
Basalt
40
42
44
SiO2
SiO2
0.4
0.8
1.2
P2O5
10 11 12 13
MgO
10.0
11.0
FeO
2.4
2.5
2.6
2.7
2.8
TiO2
Lower tuff sequence
Mid-scoriaceous unit
Upper tuff sequence
Upper scoriaceous unit
Scoria mounds
Lava
10.5
11.5
Olivine
Clinopyroxene
5%
2
5
%
10
%
10
%
2
5%
10
%
5%
11
12
13
14
15
10 11 12 13
MgO
98
Al2O3
10 11 12 13
MgO
10 11 12 13
MgO
9
10
11
12
CaO
Crater Hill
10%
2
5%
5%
Fig. 4. Major element geochemistry of juvenile clasts, scoria and lava from Motukorea vs. MgO, all in wt%. A total alkali vs. silica plot shows the variety of basaltic rock types (after
Cox et al., 1979). Samples from Crater Hill volcano, also in the AVF (Fig. 1A), are shown for comparison (Smith et al., 2008), see discussion. Arrow denotes progression of the Crater
Hill eruptive sequence. Vectors for crystallisation of olivine and Ti-augite (percentages refer to amount of crystallisation) are shown (calculated using compositions from Deer et al.
(1966) and Tschegg et al. (2011) respectively).
366 L.E. McGee et al. / Lithos 155 (2012) 360–374

et al. (2004) (
206
Pb/
204
Pb=18.3435 ± 17,
207
Pb/
204
Pb=15.5619± 16,
208
Pb/
204
Pb= 38.2784 ± 50). Error bars are shown in Figs. 7 and 8 for
the 2SD of replicate analyses of JB-2. Pb isotope standard data are
presented in the Supplementary Data (sheet 3).
4. Geochemistry
Analyses from representative samples (all those analysed for Pb
isotopes) are presented in Table 1. The full dataset is presented in
the Supplementary Data.
4.1. Major elements
Motukorea rocks range from basalt, through basanite to nephelin-
ite, with low SiO
2
values (39–44 wt.%), and total alkali (Na
2
O+K
2
O)
values that range from 4 to 6 wt.% (Fig. 4). Samples from the tuff se-
quence have higher total alkalis and lower SiO
2
than the scoria
mounds and lava, and samples from the USU plot between these
two groups (Fig. 4). MgO varies from 9.5 to 13.5 wt.%, with samples
from the tuff sequence displaying the lowest values. Generally, SiO
2
increases and TiO
2
, FeO and P
2
O
5
decrease with increasing MgO
through the sequence (Fig. 4). CaO concentrations are constant
throughout the eruptive sequence and show no correlation with
MgO, and Al
2
O
3
contents are higher in the magmatic units than in
the tuff sequence. Limited internal trends within individual units
are observed; this is particularly striking in the tuff sequence where
the data form a broad cluster with no trend in all the major elements.
Within the scoria sequence and lava flows there is a slight correlation
between MgO and FeO, and MgO and P
2
O
5
. Samples from the
mid-scoriaceous unit show more variation in the major elements
than other units, particularly in SiO
2
and Al
2
O
3
. As in SiO
2
vs. total al-
kalis, samples from the USU lie between samples from the tuff se-
quence and those from the scoria mounds and lavas in all major
elements.
4.2. Trace elements
As with the major elements, within-unit correlations of trace ele-
ments with MgO are minimal. However, the sample suite as a whole
displays significant trends with MgO, caused by differences between
the bulk compositions of each eruptive unit. Samples from the tuff se-
quence are lower in transition metals and higher in incompatible el-
ements than the scoria mounds and lavas; samples from the USU
span the range between these units. All transition metals show
150
200
250
300
Ni
30
40
50
10 11 12 13 14
Nb/U
MgO
500
700
900
1100
Sr
4
6
8
10
Th
150
200
250
300
350
400
Zr
250
300
350
400
450
500
Cr
0.18
0.19
0.20
0.21
9 10111213
Sm/Nd
MgO
3.0
3.4
3.8
4.2
(La/Sm)N
1
0%
5%
1
0%
2
5%
5
%
Ol
C
p
x
Fig. 5. Trace element geochemistry of juvenile clasts, scoria and lava from Motukorea vs. MgO. Trace elements in ppm, MgO in wt%. Symbols as in Fig. 4. Cr, Ni and Zr values from
XRF. Vectors for crystallisation of olivine and Ti-augite are shown for Cr and Ni (references as for Fig. 4).
367L.E. McGee et al. / Lithos 155 (2012) 360–374

positive trends with MgO (Fig. 5); the trend through the sequence is
extremely linear in Ni but more kinked in Cr and Sc (Sc not shown)
showing a clearer separation between the phreatomagmatic (tuff)
and magmatic (scoria and lava) units. Incompatible elements all dis-
play a negative trend with MgO. The data for the tuff sequence typi-
cally form a cluster in all but the transition metals. Samples from
the lava flows and scoria mounds typically show a more restricted
range of values than samples from the USU and tuff sequence
(Fig. 5); however, unlike the tuff sequence, the scoria and lava display
slight correlations with MgO. Variations in trace element ratios are
seen through the sequence: (La/Sm)
N
is higher in the tuff sequence
(4–4.2), whilst the lava and scoria mounds show more variation at
lower (La/Sm)
N
(3.15–3.6), but higher Sm/Nd (0.20–0.21) compared
to the tuff (0.18–0.19). A positive trend from the tuff to the lava is
seen in Nb/U with values ranging from 34.4 to 54.9.
Rare earth element (REE) profiles (normalised to chondrite after
McDonough and Sun, 1995) for the tuff sequence, MSU, USU, scoria
mounds and lava all show a pattern of light-REE (LREE) enrichment
(Fig. 6A); the tuff sequence shows the steepest profile with the
mid-scoriaceous unit lying in the middle of this field, and the lava
the flattest. The eruptive units are most similar in the heavy-REE
(HREE), with the tuff and USU having very similar values in Er-Lu.
The sequence shows a larger range of values in the LREE with a sepa-
ration observed between the USU and the scoria mounds.
On a multi-element plot (normalised to primitive mantle after
McDonough and Sun, 1995) all units show a similar profile peaking at
Nb and Ta followed by an overall decrease in enrichment (Fig. 6B), typ-
ical of ocean island basalts (Sun and McDonough, 1989). All units show
a prominent negative K anomaly (deepest in the tuff, MSU and USU
samples), and less prominent negative Zr, Hf and U anomalies. The
mid-scoriaceous unit shows both positive and negative Sr anomalies
and also a large range in Rb and Ba (due to outliers amongst these
samples –see Supplementary Data). A slight positive Sr anomaly is
also seen in the upper scoriaceous unit.
4.3. Sr–Nd–Pb isotopes
There are six Sr–Nd isotope analyses available from Motukorea
published by Huang et al. (1997), and these show very little variation:
87
Sr/
86
Sr ranges from 0.702790 to 0.702930, and
143
Nd/
144
Nd from
0.512970 to 0.51300 (not shown). Pb isotopes analysed for the current
1
10
100
1000
Rb
Ba
Th
U Nb
TaK La
Ce
Pr Sr
Nd
Zr
HfSm
Eu
Gd
Tb
Dy
Ho
ErTm
Yb
Lu
Sample/Primitive Mantle
Tuff
MSU
USU
Scoria mounds
Lava
1
10
100
1000
La Ce Pr Nd Sm Eu Gd Tb Dy Ho Er Tm Yb Lu
Sample/Chondrite
OIB
A
B
Fig. 6. (A) Rare earth element (REE) plot normalised to chondrite (after McDonough
and Sun, 1995). (B) Multi-element plot normalised to primitive mantle (after
McDonough and Sun, 1995). Fields of samples are shown from the tuff sequence,
mid-scoriaceous unit (MSU), upper scoriaceous unit (USU), scoria mounds and lava
flows. Units become progressively less enriched in LILE and LREE from the eruption
of tuff to the emission of lavas. Black line in (B) (OIB= Ocean Island Basalt) taken
from Sun and McDonough, 1989.
38.80
38.85
38.90
38.95
19.15 19.20 19.25 19.30 19.35 19.40
15.45
15.55
15.65
15.75
38.0
38.5
39.0
39.5
40.0
18.0 18.5 19.0 19.5 20.0 20.5
15.59
15.60
15.61
207Pb/204Pb
206Pb/204Pb
208Pb/204Pb
207Pb/204Pb
HIMU
Pacific
MORB
206Pb/204Pb
208Pb/204Pb
Lower tuff
MSU
Upper tuff
USU
Scoria
Lava
AVF
AVF
Lithospheric
mantle
Pacific
MORB
HIMU
Lithospheric
mantle
NHRL
NHRL
NHRL
NHRL
2SD
Fig. 7.
208
Pb/
204
Pb,
207
Pb/
204
Pb and
206
Pb/
204
Pb for tuff bombs, scoria and lava from Motukorea. Pacific MORB and HIMU (Cook-Austral islands and St Helena) data from a compi-
lation by Stracke et al. (2003). NHRL =Northern Hemisphere Reference Line, calculated after Hart (1984). 2SD error bar based on replicate analyses of JB-2 is shown for
207
Pb/
204
Pb.
Error bars for
206
Pb/
204
Pb are similar to symbol size. Isotopic data from the AVF (McGee et al., in review) are shown as a labelled pale grey field. Data from Group B and C dykes from
Marie Byrd Land, Antarctica (Storey et al., 1999) which are thought to represent lithospheric mantle beneath New Zealand are plotted and labelled ‘Lithospheric mantle’(Hoernle et
al., 2006). A triangular pattern is observed in the Motukorea data which is particularly prominent in
207
Pb/
204
Pb vs.
206
Pb/
204
Pb space.
368 L.E. McGee et al. / Lithos 155 (2012) 360–374

study display larger variations (Fig. 7):
206
Pb/
204
Pb ratios range be-
tween 19.158 and 19.360,
207
Pb/
204
Pb between 15.589 and 15.605,
and
208
Pb/
204
Pb between 38.800 and 38.958. This places the data
between the Pacific MORB and HIMU fields, closest to Pacific MORB,
similar to other data from the AVF (McGee et al., in review)(Fig. 7).
As with the whole field data, samples from within the Motukorea se-
quence occupy a triangular area in
206
Pb/
204
Pb vs.
207
Pb/
204
Pb and an
essentially linear trend in
206
Pb/
204
Pb vs.
208
Pb/
204
Pb (Fig. 7). The
lava and scoria mounds are separated in Pb-isotopic space from the
tuff and mid- and upper scoriaceous units, with the former having
lower
207
Pb/
204
Pb and extending to lower
206
Pb/
204
Pb and
208
Pb/
204
Pb.
4.4. Compositional variation with stratigraphic height
The well-constrainederuption sequence at Motukorea allows exam-
ination of geochemical changes with the progression of eruption. In
Fig. 8 the scoria mounds and lava are placed on top of the tuff sequence
and upper scoriaceous unit and given an arbitrary height, as this is
where they are known to occur temporally in the eruptive sequence
(see Section 2). A gradual increase in Mg# (magnesium number, calcu-
lated as mole percent Mg/Mg+ Fe x100, with FeO calculated from
Fe
2
O
3
tot
as FeO = Fe
2
O
3
tot
/1.31134) is seen through the tuff sequence,
followed bya more dramatic increase in the upper scoriaceous unit, sco-
ria mounds and lava (Fig. 8). A gradual decrease in incompatible ele-
ments such as La (Fig. 8), Zr and Th (not shown) is observed, with a
more dramatic decrease in the magmatic units. Some element ratios
such as Sm/Nd (Fig. 8) and La/Sm (not shown) show very little variation
through the tuff sequence followed by a change in the scoria and lava.
The MSU typically shows a large rangein values compared to the overall
variation in the tuff sequence; for example Mg# varies from 60 to 63 in
the lower and upper tuff, but the mid-scoriaceous unit varies from 62 to
64 (Fig. 8). This unit also displays a larger range of variation in fluid mo-
bile elements (Rb, Ba, Sr, U) compared to neighbouring fluid immobile
elements (Th and Nb) on Fig. 6B. High variability of these elements
has been linked to post-eruptive alteration (Schiano et al., 1993). This
may explain the variation seen in this particularlayer, as it is highly po-
rous compared to the upper and lower tuff sequences above and below
it. Systematic changes with stratigraphic height are alsoobservedfor Pb
isotope data.
208
Pb/
204
Pb (and
206
Pb/
204
Pb, not shown) decrease from
the start of the eruption until c.12 m height (i.e. in the upper tuff;
note that the upper tuff sample which plots with the upper scoriaceous
unit samples was taken from the layer directly below the latter). The
upper scoriaceous unit has a slightly more radiogenic Pb isotope com-
position compared to the start of the tuff eruption, and this decreases
in the scoria mounds and lava (Fig. 8). A pronounced decrease
in
207
Pb/
204
Pb is observed through the eruption sequence, with similar
values displayed through the tuff and most of the USU samples, and
distinctly lower values in the scoria mound and lava samples. The
cause of these systematic compositional differences and the decoupling
of the
206
Pb/
204
Pb and
208
Pb/
204
Pb ratios from the
207
Pb/
204
Pb ratio
merits further discussion.
5. Discussion
The separation of the samplesfrom the tuff sequence and thosefrom
the scoria mounds and lavas in major and trace element and Pb-isotopic
compositions, together with the non-linear nature of the sequence
in bivariate plots (Figs. 4 and 5), suggests that the dominantly
phreatomagmatic and dominantly magmatic phases of the eruption
were the products of two different magma batches, with the upper
Mg#
Sm/Nd
0
5
10
15
20
25
Height (m)
Lower tuff
MSU
Upper tuff
USU
Scoria
Lava
0
5
10
15
20
25
Height (m)
60 62 64 66 68 70 30 50 70 90
La
208Pb/204Pb
0.18 0. 19 0.20 0.21 38.8 38.9 39.0 15.5 9 15.60 15.61
207Pb/204Pb
2SD
2SD
Fig. 8. Selected element and isotope plots vs. stratigraphic height of sample (in metres) showing change in chemistry up-sequence. MSU =Mid-scoriaceous unit, USU =upper sco-
riaceous unit. The scoria mounds and lava are assigned an arbitrary stratigraphic position at the top of the sequence. In Mg#, La and Sm/Nd a gradual change in chemistry is seen
throughout the tuff sequence with a more dramatic change between tuff and magmatic phases (i.e. scoria and lava). The mid-scoriaceous unit (S17, see logs in Fig. 3) often shows a
large range in values for samples taken from this layer most likely due to post-depositional alteration.
208
Pb/
204
Pb ratios up-sequence decrease throughout the tuff sequence, then
return to a similar starting composition in the magmatic phase before again decreasing; however,
207
Pb/
204
Pb ratios decrease from the tuff to the magmatic phases. This is sugges-
tive of early depletion of a radiogenic component and the presence of two melt batches. 2SD error bars are shown for Pb isotopes.
369L.E. McGee et al. / Lithos 155 (2012) 360–374

scoriaceous unit having compositions between these two groups
suggesting mixing. The triangular pattern observed in Pb-isotopic
space implicates several sources in the magma genesis. In the following
discussion the nature and extent of the magmatic processes affecting
both magma batches, together with the characterisation of the sources
involved, reveals a far more complex melting scenario than may have
first been assumed from the island's monogenetic appearance ex-
pressed in its physical volcanology (Section 2 and Fig. 3).
5.1. The limited effect of fractional crystallisation
Samples representing the eruptive products of Motukorea have
Mg#s of 60–70 and have high MgO contents (9.5–13.5 wt.%) which in-
dicate that they are near-primary melts. The lack of trends with MgO
within eruptive units in the major elements and some of the trace ele-
ments (Figs. 4 and 5), a feature that is particularly prominent in the
tuff sequence, suggests that only minimal fractional crystallisation has
taken place during magma ascent. Olivine and clinopyroxene (Cpx –
titaniferous augite) are observed petrographically in Motukorea rocks;
thus, fractional crystallisation modelling of these minerals was under-
taken in order to see what effect this process has had on compositions
(plotted as vectors on Figs. 4 and 5). Although small amounts of frac-
tional crystallisation of olivine+Cpx in an approximately 50:50 ratio
could explain the trends seen in MgO vs. SiO
2
and FeO for the whole
suite of samples, this cannot explain the trends in MgO vs. TiO
2
,Al
2
O
3
,
Cr and Ni (Figs. 4 and 5). It is possible that the lava samples have been
affected by very small (b3%) amounts of fractional crystallisation of ol-
ivine; minor crystallisation in these samples is consistent with their
near-primitive nature (Mg# 70). The low MgO nature of the tuff sam-
ples cannot be attributed to fractionation from magmas that produced
the more primitive (higher MgO) scoria and lava samples, and this re-
quires further explanation.
These findings are in contrast with the model of Smith et al. (2008)
based on another well-exposed tuff-scoria-lava sequence in the AVF
(Crater Hill –compositional field and progression of eruption are
shown in Fig. 4, location marked on Fig. 1A). The entire volcanic se-
quence at Crater Hill was sampled at high resolution and modelled as
being controlled by deep-seated fractionation of high pressure (high
Al
2
O
3
) Cpx. The positive trend in MgO vs. Al
2
O
3
and the lack of trend
in MgO vs. CaO observed in the Motukorea sequence contrasts with
those displayed by the Crater Hill samples (Fig. 4). A notable similarity,
however, is that both eruptions began with the phreatomagmatic erup-
tion of low MgO basalt and progressed to effusive eruption of more
mafic lavas. The lack of a fractional crystallisation trend in the first
magmas to be produced at Motukorea is significant as it suggests that
they did not stall or pond en route and therefore erupted as virtually
unmodified melts of their mantle source. Although the magmas giving
rise to the lavas experienced minor amounts of shallow fractionation,
this is minimal, suggesting all samples represent near-primary melts.
5.2. Source compositions and involvement during progression of melting
The triangular pattern observed in the Pb-isotope compositions
of samples from the Motukorea eruptive sequence (Fig. 7)reveals
the involvement of three source components in the petrogenesis of
the magmas: a low
206
Pb/
204
Pb component plotting to the left of the
Northern Hemisphere Reference Line (NHRL, after Hart, 1984)inthe
206
Pb/
204
Pb vs.
208
Pb/
204
Pb diagram (represented mostly by samples
from the upper tuff sequence), a high
207
Pb/
204
Pb and
208
Pb/
204
Pb com-
ponent (represented by samples from the lower tuff sequence and
some upper scoriaceous unit samples) and a low
207
Pb/
204
Pb compo-
nent plotting close to the NHRL in both diagrams (represented by sam-
ples from the scoria mounds and some upper scoriaceous unit samples)
(Fig. 7). The composition of the low
206
Pb/
204
Pb end-member has
already been interpreted in other studies of New Zealand intraplate
volcanic fields as evidence for a subduction-modified lithospheric
component (Fig. 7)(Hoernle et al., 2006; Timm et al., 2010). This assim-
ilation of melts has been identified as occurring to varying degrees on
the scale of the whole volcanic field (McGee et al., in review). Because
Pb-isotopic compositions of both the tuff sequence and the scorias and
lavas trend towards this region in Pb-isotopic space it is suggested
that, as the magmas ascended, they assimilated melts from the overlying
subduction-metasomatised lithosphere.
The difference in Pb-isotopic composition between the beginning
of the tuff eruption and the eruption of the scoria and lavas (most
noticeable in stratigraphic heightvs.
207
Pb/
204
Pb, Fig. 8) implies a differ-
ence in the source region for the melt that generated the tuff samples
compared to those that generated the scoria and lava flows. The higher
207
Pb/
204
Pb in the lower tuff samples and their relatively low Sm/Nd
ratios (Figs. 5 and 8) is suggestive ofmelting in the presence of recycled
material which is older, and more radiogenic, than the ambient
asthenospheric mantle (DePaolo and Wasserburg, 1976). The melting
of garnet pyroxenite or eclogite veins (commonly related to ancient
subduction) is known to cause Pb-isotopic ratios to be more radiogenic
due to higher U/Pb ratios (e.g. Pilet et al., 2008; Thirlwall, 1997;
Willbold and Stracke, 2006). It is likely that such material was incorpo-
rated into the mantle in the Mesozoic subduction episode of the North
Island of New Zealand (e.g. Cook et al., 2005), explaining the somewhat
elevated Pb-isotopic values, whilst not displaying truly HIMU values
(Thirlwall, 1997; Vidal, 1992); this may explain the partial decoupling
of
206
Pb/
204
Pb and
208
Pb/
204
Pb from
207
Pb/
204
Pb (Fig. 8).
The preferential melting of eclogite veins compared to a dominantly
peridotitic mantle at low degrees of partial melting has been suggested
as the cause for magmas to be silica-undersaturated (e.g. Hirschmann et
al., 2003; Kogiso et al., 2003), resulting in the eruption of nephelinite
compositions, as is noted in the tuff sequence of Motukorea (Fig. 4). Un-
dersaturated magma compositions are a striking feature of small-
volume basaltic systems (e.g. the South Auckland volcanic field (Cook
et al., 2005), the Wudalianchi volcanic field (Hsu and Chen, 1998), the
Newer Volcanic Province (Demidjuk et al., 2007), the Eifel volcanic
field, Germany (Haase et al., 2004) and the Harrat Ash Shaam volcanic
field, Jordan (Shaw et al., 2003)) suggesting that a discontinuous,
eclogitic component may be a common feature of such fields. The
lower
207
Pb/
204
Pb ratio of samples from the scoriamounds and lava sug-
gests that they were produced from melting of the residual peridotitic
mineralogy remaining after extraction of the silica-undersaturated
magma.
The steepness of the REE plot for all eruptive units (Fig. 6A) is indic-
ative of residual garnet in the source (e.g. Hoernle and Schminke, 1993).
This is evidence for the magma being sourced from depths of ≥80 km
(e.g. McKenzie and O'Nions, 1991). The
230
Th-excess observed by
Huang et al. (1997) in four samples from Motukorea from all parts
of the sequence also supports melting in the garnet stability field,
based on the incompatibility of Th in garnet (e.g. Elliott, 1997; Peate
and Hawkesworth, 2005). The AVF contains some of the highest
(
230
Th/
232
Th) ratios for continental basalts (up to 1.38, McGee et al.,
2011, in review), suggesting that melting initiates at the bottom of a
long column (e.g. Peate and Hawkesworth, 2005). We therefore suggest
that the sequence of deposits erupted at Motukorea involved two
magma batches originating from a heterogeneous garnet-bearing as-
thenospheric source that was progressively depleted due to the exhaus-
tion of an eclogitic component. These magmas then mixed with melts
from the overlying subduction-metasomatised lithosphere, although
the relatively un-radiogenic nature of
87
Sr/
86
Sr ratios from Motukorea
(see Section 4.3) and the similarity of the multi-element pattern for
all samples suggest that this interaction was not extensive.
5.3. The nature and behaviour of the source beneath Motukorea
Using the points discussed above, we model the mantle processes that
produced the Motukorea sequence (Fig. 9). In (Gd/Yb)
N
vs. (La/Yb)
N
space samples plot in three distinct groups, each with a different trend:
370 L.E. McGee et al. / Lithos 155 (2012) 360–374

1: the tuff sequence, 2: the scoria and lava samples, and 3: samples from
mainly the upper scoriaceous unit (USU). These trends correspond to
three distinct mixing events involving the three sources identified in
Pb-isotopic space. (Gd/Yb)
N
vs. (La/Yb)
N
is used as a proxy for both de-
gree of melting as well as the nature and proportion of the aluminous
mineral phase in the mantle source (i.e. spinel vs. garnet). As we have
shown that the magmas have not been significantly affected by fractional
crystallisation processes en route to the surface (see Section 5.1)the
compositions can be viewed as essentially unmodified, and therefore
appropriate for source composition and melting modelling.
Three source components related to those identified in Pb-isotopic
space (Fig. 7) are modelled in Fig. 9 using starting compositions and
mineral modes based on the three components identified on the scale
of the whole volcanic field in McGee et al. (in review), and calculated
using the batch melting equation taking into account melting propor-
tions of each mineral (after Thirlwall et al., 1994). Parameters of the
model are given in Table 2. We model the melting of eclogite veins
(source A, Fig. 9) within a fertile garnet peridotite mantle (B) (both as-
thenospheric) and a spinel-bearing lithospheric source (C). Source A
contains a high proportion of clinopyroxene and garnet to reflect a typ-
ical eclogitic mineralogy, source B contains 5% garnet, has a mineralogy
typical of a fertile peridotite (Table 2) and has a composition slightly
more enriched than primitive mantle, and source C contains 4% spinel
and has a composition more depleted than primitive mantle to simulate
a less fertile peridotite. Source B is intended to reflect the mantle sur-
rounding the eclogite domains once melting has exhausted the latter.
Mixing lines are plotted between the lithospheric source (C) and both
asthenospheric sources (A and B) to simulate the incorporation of lith-
ospheric melts during ascent of asthenospheric magmas.
Afirst-order observation in the trace element data is that there is a
two-fold decrease in concentration from the tuff samples to the lava
and scoria samples in the incompatible elements (e.g. Zr, Sr, Th Fig. 5,
also the LREEs (Fig. 6A) and Nb, not shown). As the effect of fractional
crystallisation has been shown to be very minimal (Section 5.1), it can
be assumed that this decrease is caused by approximately doubling
the degreeof partial melting of the sources involved. Generally smallde-
grees of melting are also evident from the LILE and LREE enriched nature
of the samples from Motukorea. At very small degrees of partial melting
(b1%) a mantle source containing veins of pyroxenitic or eclogitic mate-
rial can generate silica-undersaturated melts due to the higher fusibility
of pyroxenite compared to peridotite (e.g. Hirschmann et al., 2003; Pilet
Lower tuff
Upper tuff
USU
Scoria
Lava
MSU
(Gd/Yb)N
1.0 1.5 2.0 2 .5 3.0 3.5 4.0 4.5 5.0
(La/Yb)N
0
5
10
15
20
25
30
35
A
C
B
Eclogite (athenosphere)
Fertile peridotite (garnet-bearing)
Lithospheric source (spinel-bearing)
Mixing (% of lithospheric melt)
0.5%
1%
A
C
B
0.75%
1%
1.5%
2%
10%
30%
50%
70%
90%
20%
40%
2%
2.5%
1% Degree of partial melting
1%
60%
80%
Fig. 9. (Gd/Yb)
N
vs. (La/Yb)
N
(normalised to primitive mantle after McDonough and Sun, 1995) for the Motukorea sequence to illustrate melting and mixing processes which create
the observed chemical trends between the phreatomagmatic and magmatic phases of the eruption. Solid lines show melting of three sources based on those modelled for the AVF in
McGee et al. (in review): Line A denotes melting of eclogite veins, Line B denotes melting of a fertile asthenospheric garnet peridotite source and Line C denotes melting of more
depleted lithospheric spinel-bearing peridotitic source. Italicised numbers in all models denote degrees of partial melting. Bold numbers on mixing lines show percentage of lith-
ospheric melt. See Section 5.3 for discussion of model and Table 2 for melt proportions, source modes, starting compositions and partition coefficients.
Table 2
Parameters used in melting model (Fig. 9). Melt compositions were calculated using the
batch melting equation taking into account the melting proportion of each mineral:
C
m
=1/Xx (1-((1-(PxX/D)) ^ (1/P)))x C
o
where C
m
=composition of the melt, X=
degree of partial melting, P=bulk melting proportion, D=bulk partition coefficient and
C
o
=source composition. Depleted and fertile peridotites are based on primitive mantle
(Hofmann, 1988), and eclogite composition is based on xenolith K91-11 from Barth et
al. (2001). Melting proportions for peridotites are taken from Thirlwall et al. (1994),
and based on these for eclogite. Partition coefficients are from
a
Adam and Green
(2006) and
b
Green et al. (2000); all others are from McKenzie and O'Nions (1991).
Lithosphere Asthenosphere
Depleted
peridotite (C)
Fertile
peridotite (B)
Eclogite (A)
Source
composition (C
o
)
La/Yb 0.94 1.47 0.44
Gd/Yb 1.31 1.19 1.03
Source mode Olivine 55% 55% 5%
Clinopyroxene 15% 17% 50%
Orthopyroxene 26% 23% 5%
Spinel/Garnet Spinel 4% Garnet 5% Garnet 40%
Melting
proportions
Olivine 10% 5% %
Clinopyroxene 50% 30% 35%
Orthopyroxene 27% 20% 15%
Spinel/Garnet Spinel 13% Garnet 45% Garnet 40%
Degree of melting (X) 0.5% 1–2% 0.75%
Partition
coefficients
La Gd Yb
Olivine 0.0004 0.0015 0.0015
Clinopyroxene 0.03
a
0.3 0.43
a
Orthopyroxene 0.0006
a
0.034
b
0.077
a
Spinel 0.0005
a
0.498 4.54
a
Garnet 0.01 0.01 0.01
371L.E. McGee et al. / Lithos 155 (2012) 360–374

et al., 2008). We model that as this melt ascends through the litho-
sphere, a very small degree of melting is generated (0.25% of source C)
and a small percentage of these melts (b25%) is incorporated into the
rising nephelinitic magmas (A–C mixing line in Fig. 9). The veins now
exhausted, the asthenospheric source is of a peridotitic composition
(source B), and there is a two-fold increase in degree of melting of this
source (c.2%). On ascent these melts also incorporate a small percentage
of lithospheric melt (mixing line B-C).
Larger degrees of melting (2%) of source A would eliminate the need
for a peridotitic source (B), as this would produce a similar trend with
lithospheric melts. However, this would require that the second, larger
degree asthenospheric melt mixes in a 50:50 ratio with a lithospheric
melt. This large addition of a subduction-metasomatised melt would
be expected to have an effecton the trace element and isotopic charac-
teristics of the resultant magmas. Volcanic centres in the AVF which
have been modelled as incorporating a large proportion of lithospheric
melt display an obvious positive Sr anomaly on a multi-element
plot, and also lie towards lower
206
Pb/
204
Pb and higher
207
Pb/
204
Pb
(McGee et al., in review). Neither of these features is evident in samples
from the scoria mounds and lavas at Motukorea (Figs. 6Band7). Note
that although samples from the upper scoriaceous unit lie along the py-
roxenitic asthenosphere-lithosphere mixing line (A–C), we attribute
their chemistry to mixing between the tuff-producing magma and
melts of the peridotitic source, instead of incorporation of c. 50% litho-
spheric melts. Samples from this unit consistently plot as a transition
between the tuff and scoria and lava compositions (Figs.4,5,8,and9)
showing that they are produced by mixing of the two magma batches.
In addition to this, the compositions of samples from this unit show
no evidence of having incorporated a substantial proportion of litho-
spheric melt, as discussed above. We therefore favour the hypothesis
that the progressive melting of an eclogitic source leads to exhaustion
of this component and eruption of nephelinitic compositions, followed
by production of a second melt batch by larger degrees of melting of a
peridotitic source which produced basanite and alkalic basalt composi-
tions. Mixing between these two melt batches produced thetransitional
compositions of the upper scoriaceous deposit. Depletion of the source
region with the progression of melting and consequent eruption is a
logical concept (e.g. Reiners, 1998; Reiners and Nelson, 1998), and we
demonstrate that this can actually be observed as a result of high reso-
lution sampling through a continuous eruption sequence.
5.4. Linking composition and volcanic stratigraphy
A schematic of the model for Motukorea is shown in Fig. 10 illustrat-
ing both the compositional and volcanological progression of the erup-
tion. The upper and lower tuff sequences are shown as being produced
from the same source and with the same melting conditions, explaining
the general similarity in composition throughout the tuff sequence (1).
The juvenile samples in these deposits are nephelinitic in composition
due to the preferential melting of eclogitic veins. As suggested by
Bryner (1991), the mid-scoriaceous unit (MSU) is likely to have formed
due to the temporary exhaustion of local groundwater (2), with the
phreatomagmatic eruption continuing after deposition of this juvenile-
clast-dominated layer presumably due to recharge of the water supply
(3). The phreatomagmatic stage of the eruption is then terminated by
the arrival at the surface of larger volume and larger degree melts
Groundwater
Crust
Asthenosphere
MSU
Mixing of
melts
Deposition
of USU
Termination of
phreatomagmatic
phase
Recharge of
primitive magma,
larger degree
of melting
Deposition of
lower tuff unit
Ascent through
subduction-
metasomatised
lithosphere
Silica-undersaturated
small degree melts
(=Nephelinites)
Gt/Sp
transition
Deposition
of MSU Drying out of
groundwater
Deposition of
upper tuff unit
Recharge of
groundwater
Completed
tuff ring
USU
Formation of
scoria cone
Rafting of scoria
by lava flows
More primitive,
less enriched
alkalic basalts
and basanites
Most primitive
lavas erupted
(4) (5) (6)
Continued
building of
scoria cone
eclogite
veins
Exhaustion of
eclogite veins
Emission of
lava from
scoria cone
vent
(1) (2) (3)
Fig. 10. Schematic model of the eruption of Motukorea illustrating the volcanological features and geochemical trends observed. MSU =mid-scoriaceous unit, USU=upper scori-
aceous unit. Gt/Sp transition =garnet/spinel transition. (1)–(3) depict the melting of eclogite veins resulting in the deposition of the tuff sequence, (4) depicts the mixing of the first
magma batch with the second, produced by larger degrees of melting leading to the deposition of the transitional USU. (5)–(6) illustrate the ascent of the second, possibly larger
magma batch and resultant formation of the scoria cone, emission of lava flows and rafting of scoriaceous material. See Section 5.4 for further discussion.
372 L.E. McGee et al. / Lithos 155 (2012) 360–374

which overwhelm the environment by increasing the magma:water
ratio, leading to the onset of magmatic activity with the eruption of
the upper scoriaceous unit (USU); within-conduit mixing explains the
transitional nature of this deposit's composition. This coincides with
the exhaustion of the eclogite source component (4). A scoria cone
begins to develop, as the eruption style becomes purely strombolian
(5); material produced is basanitic to basaltic in composition. As the
eruption progresses, lava is emitted from the main vent and rafts some
of the scoriaceous material away from the cone to form the scoria
mounds, whilst the cone continues to form (6).
It is possible that the magma flux was higher in the second magma
batch; the evidence for this is: 1) the two-fold increase in degree of melt-
ing in this second magma batch (Section 5.3,Fig. 9), 2) the larger volume
of the resultant eruptive products (scoria cone and lava flows)
(Section 2) and 3) the abrupt change from phreatomagmatic to magmat-
ic volcanism which is indicative of a higher magma:water ratio. Our find-
ings support the hypothesis of Strong and Wolff (2003) that eruptions at
monogenetic volcanoes may occur in pairs, with the first magma being a
‘path-finder’for a second, larger (in the case of Motukorea) magma
batch.
The mixing of the first and second melt batch, inferred from the
transitional nature of the upper scoriaceous unit, is significant in
that it shows that the tail of the small-degree nephelinitic magma
was followed immediately by the head of the second more primitive
melt (panel 4, Fig. 10), with no hiatus in the volcanic record or in the
chemical trends observed (Figs. 3–5, 8, and 9). This is evidence that a
seemingly monogenetic eruption can be composed of more than one
magma batch, i.e. be’compositionally polygenetic’whilst remaining
‘superficially monogenetic’, and also suggests that the conduit re-
mains open between the two melt batches despite the very small vol-
umes involved.
6. Conclusions and implications for monogenetic volcanism
Major and trace element and Pb-isotopic compositions of
stratigraphically controlled samples have revealed details about the
sources and melting conditions of a complete monogenetic sequence.
Systematic chemical changes through the phreatomagmatic phase of
the eruption to the purelymagmatic phase indicates a far morecomplex
melting scenario than the term ‘monogenetic’implies. We have shown
that the magma represented by the early tuff sequence and later scoria
experienced no fractional crystallisation, whilst that represented by the
lava experienced only very minor crystallisation of olivine. Modelling
shows that juvenile material in the tuff ring was produced from very
small degrees of melting of eclogite veins within the asthenosphere,
and that this component was exhausted during melting. A second
magma batch was produced from the remaining garnet peridotite
source, at degrees of melting twice as large as in the initial magma
batch. A compositionally transitional deposit shows that there was no
hiatus in the ascent of these melt batches. The magma recharge event
may have been accompanied by an increase in magma flux, evidenced
by the transition to magmatic eruptions and the larger volume of erup-
tive products in the magmatic phase.
Our model is in good agreement with the findings of past studies
that have illustrated that monogenetic eruptions can show considerable
compositional differences through sequences and between phases of
the eruption (Brenna et al., 2010, 2011; Erlund et al., 2010; Siebe et
al., 2004; Strong and Wolff, 2003). Although the term ‘monogenetic’
can be applied to the volcanology of such sequences (in that they define
a continuous eruptive episode), this label does not adequately describe
the complex melting and mixing processes within mantle sources
which are often reflected in the geochemistry. The detailed analysis of
a complete small-scale volcanic sequence such as at Motukorea allows
us to document the variations in melting conditions and changes that
the magmatic system undergoes during the progression of a monoge-
netic eruption, from start to finish, and has also shown that mantle
components identifiable on the scale of a volcanic field can also be ob-
served on the scale of a single eruption.
Acknowledgements
We thank the New Zealand Department of Conservation for access
and transportation to Motukorea. Lucas Hogan, Aleksandra Zawalna-
Geer, Gabor Kereszturi and Javier Agustin-Flores are thanked for field-
work assistance and we thank Gabor Kereszturi for the volume esti-
mates. We benefitted from field discussions with Bruce Hayward, who
we also thank for providing the aerial photograph of Motukorea. We
thank John Wilmshurst for XRF analyses and Charlotte Allen at the
Australian National University for assistance with LA-ICP-MS analyses.
Steve Blake and Terry Plank are thanked for constructive reviews of
this work. This project was funded by the Determining Volcanic Risk in
Auckland (DEVORA) project as part of LEM's PhD thesis. JML gratefully
acknowledges support from the New Zealand Earthquake Commission.
Appendix A. Supplementary data
Supplementary data to this article can be found online at http://
dx.doi.org/10.1016/j.lithos.2012.09.012.
References
Adam, J., Green, T., 2006. Trace element partitioning between mica- and amphibole-
bearing garnet lherzolite and hydrous basanitic melt: 1. Experimental results and
the investigation of controls on partitioning behaviour. Contributions to Mineralogy
and Petrology 152, 1–17.
Allen, S.R., Smith, I.E., 1994. Eruption styles and volcanic hazard in the Auckland Volcanic
Field, New Zealand. Geoscience Reports of Shizuoka University 20, 5–14.
Allen, S.R., Bryner, V.F., Smith, I.E.M., Ballance, P.F., 1996. Facies analysis of pyroclastic
deposits within basaltic tuff-rings of the Auckland volcanic field, New Zealand.
New Zealand Journal of Geology and Geophysics 39, 309–327.
Baker, J., Peate, D., Waight, T., Meyzen, C., 2004. Pb isotopic analysis of standards and
samples using a
207
Pb–
204
Pb double spike and thallium to correct for mass bias
with a double-focusing MC-ICP-MS. Chemical Geology 211, 275–303.
Barth, M.G., Rudnick, R.L., Horn, I., McDonough, W.F., Spicuzza, M.J., Valley, J.W.,
Haggerty, S.E., 2001. Geochemistry of xenolithic eclogites from West Africa, part
I: a link between low MgO eclogites and archean crust formation. Geochimica et
Cosmochimica Acta 65 (9), 1499–1527.
Blondes, M.S., Reiners, P.W., Ducea, M.N., Singer, B.S., Chesley, J., 2008. Temporal-
compositional trends over short and long time-scales in basalts of the Big Pine
Volcanic Field, California. Earth and Planetary Science Letters 269, 140–154.
Bradshaw,T.K., Smith, E.I., 1994. Polygenetic Quaternary volcanismat Crater Flat, Nevada.
Journal of Volcanology and Geothermal Research 63, 165–182.
Brenna, M., Cronin, S., Smith, I., Sohn, Y., Németh, K., 2010. Mechanisms driving
polymagmatic activity at a monogenetic volcano, Udo, Jeju Island, South Korea.
Contributions to Mineralogy and Petrology 160, 931–950.
Brenna, M., Cronin, S.J., Németh, K., Smith, I.E.M., Sohn, Y.K., 2011. The influence of
magma plumbing complexity on monogenetic eruptions, Jeju Island, Korea. Terra
Nova 23, 70–75.
Bryner, V., 1991. Motukorea: the evolution of an eruption centre in the Auckland Vol-
canic Field. MSc: University of Auckland.
Cebriá, J.M., Martiny, B.M., López-Ruiz, J., Morán-Zenteno, D.J., 2011. The Parícutin calc-
alkaline lavas: newgeochemicaland petrogenetic modellingconstraints on the crustal
assimilation process. Journal of Volcanology and Geothermal Research 201, 113–125.
Connor, C.B., Conway, F.M., 2000. Basaltic Volcanic Fields. In: Sigurdsson, H. (Ed.),
Encyclopedia of Volcanoes. Academic Press, San Diego, pp. 331–343.
Cook, C., Briggs, R.M., Smith, I.E.M., Maas, R., 2005. Petrology and geochemistry of intra-
plate basalts in the South Auckland Volcanic Field, New Zealand: evidence for two
coeval magma suites from distinct sources. Journal of Petrology 46, 473–503.
Cox, K.G., Bell, J.D., Pankhurst, R.J., 1979. The interpretation of igneous rocks. Unwin
Hyman, London, p. 450.
Deer, W.A., Howie, R.A., Zussman, J., 1966. An introduction to the rock forming minerals.
Longman . 528pp.
Demidjuk, Z., Turner, S., Sandiford, M., George, R., Foden, J., Etheridge, M., 2007. U-series
isotope and geodynamic constraints on mantlemelting processes beneath the Newer
Volcanic Province in South Australia. Earth and Planetary Science Letters 261,
517–533.
DePaolo, D.J., Wasserburg, G.J., 1976. Nd isotopic variations and petrogenetic models.
Geophysical Research Letters 3, 249–252.
Edbrooke, S.W., 2001. Geology of the Auckland area. Institute of Geological and Nuclear
Sciences Ltd., Lower Hutt, New Zealand.
Eggins, S.M., Rudnick, R.L., McDonough, W.F., 1998. The composition of peridotites and
their minerals: a laser-ablation ICP-MS study. Earth and Planetary Science Letters
154, 53–71.
Elliott, T., 1997. Fractionation of U and Th during mantle melting: a reprise. Chemical
Geology 139, 165–183.
373L.E. McGee et al. / Lithos 155 (2012) 360–374

Erlund, E.J., Cashman, K.V., Wallace, P.J., Pioli, L., Rosi, M., Johnson, E., Granados, H.D.,
2010. Compositional evolution of magma from Parícutin Volcano, Mexico: the
tephra record. Journal of Volcanology and Geothermal Research 197, 167–187.
Garcia, M.O., Pietruszka, A.J., Rhodes, J.M., Swanson, K., 2000. Magmatic processes during
the prolonged Pu'u’O'o Eruption of Kilauea Volcano, Hawaii. Journal of Petrology 41,
967–990.
Green, T.H., Blundy, J.D., Adam, J., Yaxley, G.M., 2000. SIMS determination of trace
element partition coefficients between garnet, clinopyroxene and hydrous basaltic
liquids at 2–7.5 GPa and 1080–1200 °C. Lithos 53 (3–4), 165–187.
Haase, K.M., Goldschmidt, B., Garbe-Schoenberg, C.-D., 2004. Petrogenesis of tertiary
continental intra-plate lavas from the Westerwald region, Germany. Journal of
Petrology 45, 883–905.
Hart, S.R., 1984. A large-scale isotope anomaly in the Southern Hemisphere mantle.
Nature 309, 753–757.
Hirschmann, M.M., Kogiso, T., Baker, M.B., Stolper, E.M., 2003. Alkalic magmas generated
by partial melting of garnet pyroxenite. Geology 31, 481–484.
Hoernle, K., Schminke, H.-U., 1993. The petrology of the tholeiites through melilite
nephelinites on Gran Canaria, Canary Islands: crystal fractionation, accumulation,
and depths of melting. Journal of Petrology 34, 573–597.
Hoernle, K., White, J.D.L., van den Bogaard, P., Hauff, F., Coombs, D.S., Werner, R., Timm,
C., Garbe-Schönberg, D., Reay, A., Cooper, A.F., 2006. Cenozoic intraplate volcanism
on New Zealand: upwelling induced by lithospheric removal. Earth and Planetary
Science Letters 248, 350–367.
Hofmann, A.W., 1988. Chemical differentiation of the Earth: the relationship between
mantle, continental crust, and oceanic crust. Earth and Planetary Science Letters
90 (3), 297–314.
Houghton, B.F., Wilson, C.J.N., Smith, I.E.M., 1999. Shallow-seated controls on styles of
explosive basaltic volcanism: a case study from New Zealand. Journal of Volcanology
and Geothermal Research 91, 97–120.
Hsu, C., Chen, J., 1998. Geochemistry of late Cenozoic basalts from Wudalianchi and
Jingpohu areas, Heilongjiang province, northeast China. Journal of Asian Earth
Science 16, 385–405.
Huang, Y.M., Hawkesworth, C., van Calsteren, P., Smith, I., Black, P., 1997. Melt genera-
tion models for the Auckland volcanic field, New Zealand: constraints from U-Th
isotopes. Earth and Planetary Science Letters 149, 67–84.
Kereszturi, G., Procter, J., Cronin, S.J., Németh, K., Bebbington, M., Lindsay, J., 2012.
LiDAR-based quantification of lava flow susceptibility in the City of Auckland
(New Zealand). Remote Sensing of Environment 125, 198–213.
Kienle, J., Kyle, P.R., Self, S., Motyka, R.J., Lorenz, V., 1980. Ukinrek Maars, Alaska, I. April
1977 eruption sequence, petrology and tectonic setting. Journal of Volcanology and
Geothermal Research 7 (1–2), 11–37.
Kogiso, T., Hirschmann, M.M., Frost, D.J., 2003. High-pressure partial melting of garnet
pyroxenite: possible mafic lithologies in the source of ocean island basalts. Earth
and Planetary Science Letters 216, 603–617.
Lindsay, J.M., Leonard, G.S., Smid, E.R.,Hayward, B.W., 2011. Ageof the Auckland Volcanic
Field: a review of existing data. New Zealand Journal of Geology and Geophysics 54,
379–401.
McDonough, W.F., Sun, S.-s, 1995. The composition of the Earth. Chemical Geology 120,
223–253.
McGee, L., Beier, C., Smith, I., Turner, S., 2011. Dynamics of melting beneath a small-scale
basaltic system: a U-Th–Ra study from Rangitoto volcano, Auckland volcanic field,
New Zealand. Contributions to Mineralogy and Petrology 162, 547–563.
McGee, L.E., Smith, I.E.M., Millet, M.-A., Handley, H.K., Lindsay, J.M., in review.
Asthenospheric control of melting processes in a monogenetic basaltic system: a
case study of the Auckland Volcanic Field, New Zealand. Journal of Petrology.
McKenzie, D., O'Nions, R.K., 1991. Partial melt distributions from inversion of rare earth
element concentrations. Journal of Petrology 32, 1021–1091.
Millet, M.-A., Doucelance, R., Schiano, P., David, K., Bosq, C., 2008. Mantle plume hetero-
geneity versus shallow-level interactions: a case study, the São Nicolau Island,
Cape Verde archipelago. Journal of Volcanology and Geothermal Research 176,
265–276.
Needham, A.J., Lindsay, J.M., Smith, I.E.M., Augustinus, P., Shane, P.A., 2011. Sequential
eruption of alkaline and sub-alkaline magmas from a small monogenetic volcano
in the Auckland Volcanic Field, New Zealand. Journal of Volcanology and Geothermal
Research 201, 126–142.
Németh, K., 2010. Monogenetic volcanic fields: origin, sedimentary record, and relation-
ship with polygenetic volcanism. In: Cañón-Tapia, E., Szakács, A. (Eds.), What Is a
Volcano? , pp. 43–67.
Peate, D.W., Hawkesworth, C.J., 2005. U series disequilibria: insights into mantle melting
and the timescales of magma differentiation. Reviews of Geophysics 43, RG1003.
http://dx.doi.org/10.1029/2004rg000154.
Pilet, S., Baker, M.B., Stolper, E.M., 2008. Metasomatized lithosphere and the origin of
alkaline lavas. Science 320, 916–919.
Reiners, P.W., 1998. Reactive melt transport in the mantle and geochemical signatures
of mantle-derived magmas. Journal of Petrology 39, 1039–1061.
Reiners, P.W., 2002. Temporal-compositional trends in intraplate basalt eruptions:
implications for mantle heterogeneity and melting processes. Geochemistry,
Geophysics, Geosystems 3. http://dx.doi.org/10.1029/2001gc000250.
Reiners, P., Nelson, B., 1998. Temporal-compositional-isotopic trends in rejuvenated-
stage magmas of Kauai, Hawaii, and implications for mantle melting processes.
Geochimica et Cosmochimica Acta 62, 2347–2368.
Riggs, N.R., Duffield, W.A., 2008. Record of complex scoria cone eruptive activity at Red
Mountain,Arizona, USA, and implications for monogenetic maficvolcanoes.Journalof
Volcanology and Geothermal Research 178, 763–776.
Schiano, P., Dupré, B., Lewin, E., 1993. Application of element concentration variability
to the study of basalt alteration (Fangataufa atoll, French Polynesia). Chemical
Geology 104, 99–124.
Shaw, J.E., Baker, J.A., Menzies, M.A., Thirlwell, M.F., Ibrahim, K.M., 2003. Petrogenesis of
the largest intraplate volcanic field on the Arabian plate (Jordan): a mixed
lithosphere–asthenosphere source activated by lithospheric extension. Journal of
Petrology 44, 1657–1679.
Siebe, C., Rodriguez-Lara, V., Schaaf, P., Abrams, M., 2004. Geochemistry, Sr–Nd isotope
composition, and tectonic setting of Holocene Pelado, Guespalapa and Chichinautzin
scoria cones, south of Mexico City. Journal of Volcanology and Geothermal Research
130, 197–226.
Smith, I.E.M., Blake, S., Wilson, C.J.N., Houghton, B.F., 2008. Deep-seated fractionation
during the rise of a small-volume basalt magma batch: Crater Hill, Auckland,
New Zealand. Contributions to Mineralogy and Petrology 155, 511–527.
Storey, B.C., Leat, P.T., Weaver, S.D., Pankhurst, R.J., Bradshaw, J.D., Kelley, S., 1999. Mantle
plumes and Antarctica-New Zealand rifting: evidence from mid-Cretaceous mafic
dykes. Journal of the Geological Society 156 (4), 659–671.
Stracke, A., Bizimis, M., Salters, V.J.M., 2003. Recycling oceanic crust: quantitative
constraints. Geochemistry, Geophysics, Geosystems 4. http://dx.doi.org/10.1029/
2001GC000223.
Strong, M., Wolff, J., 2003. Compositional variations within scoria cones. Geology 31,
143–146.
Sun, S.-s, McDonough, W.F., 1989. Chemical and isotopic systematics of oceanic basalts:
implications for mantle composition and processes. Geological Society, London,
Special Publications 42, 313–345.
Thirlwall, M.F., 1997. Pb isotopic and elemental evidence for OIB derivation from young
HIMU mantle. Chemical Geology 139 (1–4), 51–74.
Thirlwall, M.F., Upton, B.G.J., Jenkins, C., 1994. Interaction between Continental Litho-
sphere and the Iceland Plume—Sr–Nd–Pb Isotope Geochemistry of Tertiary Basalts,
NE Greenland. Journal of Petrology 35, 839–879.
Timm, C., Hoernle, K., Werner, R., Hauff, F., den Bogaard, P.v, White, J., Mortimer, N.,
Garbe-Schönberg, D., 2010. Temporal and geochemical evolution of the Cenozoic
intraplate volcanism of Zealandia. Earth-Science Reviews 98, 38–64.
Tschegg, C., Ntaflos, T., Akinin, V.V., 2011. Polybaric petrogenesis of Neogene alkaline
magmas in an extensional tectonic environment: Viliga Volcanic Field, northeast
Russia. Lithos 122 (1–2), 13–24.
Valentine, G.A., Gregg, T.K.P., 2008. Continental basaltic volcanoes—processes and
problems. Journal of Volcanology and Geothermal Research 177, 857–873.
Valentine, G.A., Hirano, N., 2010. Mechanisms of low-flux intraplate volcanic fields -
Basin and Range (North America) and northwest Pacific Ocean. Geological Society
of America 38, 55–58.
Valentine, G.A., Keating, G.N., 2007. Eruptive styles and inferences about plumbing
systems at Hidden Cone and Little Black Peak scoria cone volcanoes (Nevada, USA).
Bulletin of Volcanology 70, 105–113.
Vidal, P., 1992. Mantle: more HIMU in the future? Geochimica et Cosmochimica Acta 56
(12), 4295–4299.
Willbold, M., Stracke, A., 2006. Trace element composition of mantle end-members:
implications for recycling of oceanic and upper and lower continental crust.
Geochemistry, Geophysics, Geosystems 7. http://dx.doi.org/10.1029/2005GC001005.
374 L.E. McGee et al. / Lithos 155 (2012) 360–374