A cosmogenic 3
He chronology of late Quaternary glacier fluctuations
in North Island, New Zealand (39

S)
Shaun R. Eaves a, b, *
, Andrew N. Mackintosh a, b
, Gisela Winckler c
, Joerg M. Schaefer c
,
Brent V. Alloway b
, Dougal B. Townsend d
a
Antarctic Research Centre, Victoria University of Wellington, P.O. Box 600, Wellington 6140, New Zealand
b
School of Geography Environment and Earth Sciences, Victoria University of Wellington, P.O. Box 600, Wellington 6140, New Zealand
c
Lamont-Doherty Earth Observatory of Columbia University P.O. Box 1000, Palisades, NY 10964, USA
d
GNS Science, P.O. Box 30-368, Lower Hutt 5040, New Zealand
a r t i c l e i n f o
Article history:
Received 24 March 2015
Received in revised form
3 November 2015
Accepted 4 November 2015
Available online 28 November 2015
Keywords:
Cosmogenic 3
He
Last Glacial Maximum
Last glacial cycle
Equilibrium line altitude
Tephrostratigraphy
New Zealand
Southern Hemisphere
a b s t r a c t
Mountain glaciers advance and retreat primarily in response to changes in climate. Establishing the
timing and magnitude of mountain glacier fluctuations from geological records can thus help to identify
the drivers and mechanisms of past climate change. In this study, we use cosmogenic 3
He surface
exposure dating and tephrochronology to constrain the timing of past glaciation on Tongariro massif in
central North Island, New Zealand (39

S). Exposure ages from moraine boulders show that valley
glaciation persisted between c. 30e18 ka, which coincides with the global Last Glacial Maximum.
Reinterpretation of moraine tephrostratigraphy, using major element geochemistry analysis, shows that
ice retreat and climatic amelioration at the last glacial termination was well underway prior to 14 ka. The
equilibrium line altitude in central North Island, during the Last Glacial Maximum, was c. 1400e1550 m
above sea level, which is c. 930e1080 m lower than present. Considering the uncertainties in the glacial
reconstruction and temperature lapse rates, we estimate that this equilibrium line altitude lowering
equates to a temperature depression of 5.6 ± 1.1

C, relative to present. Our mapping and surface
exposure dating also show evidence for an earlier period of glaciation, of similar magnitude to the Last
Glacial Maximum, which culminated prior to 57 ka, probably during Marine Isotope Stage 4. Good
agreement between the timing and magnitude of glacier fluctuations in central North Island and the
Southern Alps indicate a response to a common climatic forcing during the last glacial cycle.
© 2015 Elsevier Ltd. All rights reserved.
1. Introduction
Explaining the drivers of Quaternary climate cycles in the
Southern Hemisphere remains an outstanding goal of palaeoclimate
research. New Zealand is one of the few locations in the
Southern Hemisphere where terrestrial palaeoclimate can be
reconstructed. Furthermore, it is ideally situated to record fluctuations
of the southern westerly winds and the oceanic subtropical
and subpolar fronts (Fig. 1), which are considered to have played an
important role in past climate dynamics (Denton et al., 2010).
Regional tectonic uplift and localised effusive volcanism have
resulted in high-elevation and high-relief topography in both the
North and South Island of New Zealand, which support contemporary
glaciers spanning latitudes from 39

S to 46

S. Empirical and
model-based analyses show contemporary glacier mass balance in
New Zealand is most sensitive to changes in atmospheric temperature
(Oerlemans, 1997; Anderson and Mackintosh, 2006;
Anderson et al., 2010; Brook et al., 2011). Steep mass balance gradients
and relatively high ice velocities mean that mass balance
changes are translated to the termini of mountain glaciers over
timescales of 101
e102
yrs (Oerlemans, 1997; Purdie et al., 2014).
Geological records of past glacier fluctuations in New Zealand
therefore represent an important proxy for past climatic change in
the southern mid-latitudes (e.g. Anderson and Mackintosh, 2006;
Schaefer et al., 2006, 2009; Kaplan et al., 2010, 2013; Putnam
et al., 2010a, 2012, 2013a, 2013b; Golledge et al., 2012; Barrows
et al., 2013; Doughty et al., 2013; Rother et al., 2014).
Existing reconstructions of late Quaternary glacier fluctuations
in New Zealand have predominantly concentrated on large outlet
* Corresponding author. Antarctic Research Centre, Victoria University of
Wellington, P.O. Box 600, Wellington 6140, New. Zealand.
E-mail address: shaun.eaves@vuw.ac.nz (S.R. Eaves).
Contents lists available at ScienceDirect
Quaternary Science Reviews
journal homepage: www.elsevier.com/locate/quascirev
http://dx.doi.org/10.1016/j.quascirev.2015.11.004
0277-3791/© 2015 Elsevier Ltd. All rights reserved.
Quaternary Science Reviews 132 (2016) 40e56
glaciers that drained the central Southern Alps (e.g. Porter, 1975;
Suggate and Almond, 2005; Schaefer et al., 2006; Barrows et al.,
2013; Putnam et al., 2013b; Rother et al., 2014). Radiocarbon and
luminescence dating of peat and loess units, interbedded with
glacial till and glaciofluvial outwash, have provided a broad chronostratigraphic
framework for this region (Suggate, 1990; Suggate
and Almond, 2005). More recently, cosmogenic surface exposure
dating of moraine boulders has provided the means to directly
constrain the timing and magnitude of past glacier fluctuations
(Schaefer et al., 2006; Barrows et al., 2013; Putnam et al., 2013a, b;
Rother et al., 2014). During the last glacial cycle, glacier length in
the Southern Alps peaked relatively early (c. 32e26 ka; Suggate and
Almond, 2005; Putnam et al., 2013b; Barrows et al., 2013; Rother
et al., 2014), in comparison to global ice volume (Clark et al.,
2009). At this time, local equilibrium line altitudes were
depressed by c. 850 m and temperatures were c. 6e6.5

C colder,
relative to present (Golledge et al., 2012). Subsequently, glaciers
fluctuated about these positions until at least c. 18 ka before rising
temperatures resulted in widespread deglaciation (Schaefer et al.,
2006; Putnam et al., 2013a, b; Rother et al., 2014). Recent work
has shown that advances of former outlet glaciers in central
Southern Alps, prior to the Last Glacial Maximum (LGM), culminated
at 42 ± 1 ka (Kelley et al., 2014), 65 ± 3 ka (Schaefer et al.,
2015) and 139 ± 11 ka (Putnam et al., 2013b). Previous research
had suggested the occurrence of significant pre-LGM glacial advances
in New Zealand (Williams, 1996; Almond et al., 2001;
Preusser et al., 2005; Sutherland et al., 2007; McCarthy et al.,
2008), but the exact timing of these events was poorly resolved
due to low chronological precision or small sample populations.
Resolving the timing and magnitude of late Quaternary glacier
fluctuations in locations outside of the central Southern Alps will
provide insight to the synchroneity and climatic gradients of past
climate change in the southern mid-latitudes. In this study, we use
cosmogenic 3
He exposure dating, tephrochronology and equilibrium
line altitude (ELA) reconstruction techniques, to provide the
first direct age constraint on the timing and magnitude of glacier
fluctuations on Tongariro massif, central North Island (39

S).
2. Setting
2.1. Regional climatic situation
Situated between 34

S and 47

S in the south west sector of the
Pacific Ocean, New Zealand spans subtropical and subpolar climes
(Fig. 1). Westerly atmospheric circulation dominates between 30

S
and 60

S and is responsible for the eastward migrating troughs and
anticyclones that define weather variability in New Zealand yearround
(Sturman and Tapper, 1996). Interception of such weather
systems by the predominantly NEeSW trending axial ranges results
in a distinct zonal precipitation gradient. Meanwhile, interannual to
interdecadal air temperature anomalies in New Zealand are
strongly influenced by upwind sea surface temperatures (Sutton
et al., 2005). The latitudinal range and high topographic relief of
the New Zealand landmass result in meridional and zonal gradients
in air temperature and precipitation respectively. Contemporary
oceanic influences on North Island, New Zealand (34e41

S) are
largely sub-tropical, predominantly originating from an eastward
flowing branch of the equatorial-sourced East Australian Current,
known as the Tasman Front, which descends the northeast coast
before continuing eastwards along the northern margin of the
Chatham Rise (Fig. 1). In contrast, South Island (40e47

S) intersects
the sub-tropical front (STF), where sub-tropical gyres and subFig.
1. (A) Contemporary general atmospheric and oceanic circulation in New Zealand and locations of sites referred to in text (background image sourced from National Institute for
Water and Atmospheric research (NIWA)). Sites: TgVC ¼ Tongariro Volcanic Centre; TaM ¼ Tararua Mts.; TsM ¼ Tasman Mts.; CSA ¼ Central Southern Alps; CV ¼ Cascade valley.
Ocean currents: EAUC ¼ East Australian Current; ECC ¼ East Cape Current; STF ¼ Sub-tropical front; SAF ¼ Sub-Antarctic Front; ACC ¼ Antarctic Circumpolar Current; (B) Hillshaded
digital elevation model (Columbus et al., 2011) of the Tongariro massif and Mt. Ruapehu, with the main study region defined (black outline). The location of proximal
tephra reference site at Taurewa and Mt. Ruapehu glacier outlines according to Keys (1988) are also shown.
S.R. Eaves et al. / Quaternary Science Reviews 132 (2016) 40e56 41
Antarctic water masses converge, representing a temperature,
salinity and nutrient boundary, which defines the northern margin
of the Southern Ocean (Sikes et al., 2009). Consequently, steep
zonal and meridional sea surface temperature (SST) gradients exist
across the New Zealand region (Uddstrom and Oien, 1999).
2.2. Study site and previous work
Tongariro massif in central North Island (39.13

S, 175.65

E)
forms part of Tongariro Volcanic Centre (TgVC; Fig. 1), which is an
area of andesitic volcanism at the southwestern end of Taupo
Volcanic Zone. Tongariro massif comprises c. 17 coalescing volcanic
vents, with ages ranging from c. 270 ka to present (Hobden et al.,
1996). The local climate, as recorded at Whakapapa village
(1097 m asl) located 8 km SSW of our study site, is characterised by
low seasonal precipitation variability, with total annual precipitation
averaging c. 2800 mm (AD1981e2010; NIWA, 2014). Monthly
mean temperatures at this station range from c. 13

C in February to
c. 3

C in July, with an annual average of 7.5

C (NIWA, 2014).
Geomorphological evidence for past glaciation on Tongariro
massif has long been recognised (Mathews, 1967; Topping, 1974),
however lack of chronological constraint has so far precluded any
palaeoclimatic interpretation. Mathews (1967) first described
conspicuous, lateral moraines in several valleys radiating from the
volcano and correlate these landforms to the late Pleistocene moraines
of the Southern Alps, based on their size and degree of
preservation. Topping and Kohn (1973) identified two rhyolitic
beds in a soil section overlying a large lateral moraine in the
Mangatepopo valley, on the western side of Tongariro massif
(Fig. 1). Using analyses of titano-magnetite assemblages, they
correlated the lower bed to the Rerewhakaaitu Tephra (c.
17.5 ± 0.5 ka BP e Lowe et al., 2013), which is sourced from Okataina
Volcanic Centre (OVC), situated c. 130 km north of Tongariro massif.
This finding was then used to stratigraphically correlate the upper
bed to the Waiohau Tephra (cf. Donoghue et al., 1995), also sourced
from the OVC, at 14.0 ± 0.2 ka (Lowe et al., 2013). On the eastern
flanks of Tongariro massif, Cronin and Neall (1997) identified both
the Waiohau and Rerewhakaaitu Tephras at the base of soils overlying
lateral moraines. These stratigraphic constraints provide
minimum ages for the underlying moraines.
On nearby Mt. Ruapehu (2797 m asl), situated c. 15 km to the
south west of Tongariro massif (Fig. 1), McArthur and Shepherd
(1990) identify large lateral moraines in several valleys, which
descend to an altitude of c. 1200 m asl. They suggest the former ice
mass existed during the Last Glacial Maximum, based on the
presence of the Kawakawa/Oruanui Tephra (25.4 ± 0.2 ka e
Vandergoes et al., 2013), which is found interbedded within
moraine-bound glaciolacustrine deposits.
2.2.1. Mangatepopo valley
The Mangatepopo valley is c. 6 km long and 1e2 km wide, with
small, westward-flowing under-fit streams draining both the
northern and southern valley margins, which converge at the valley
mouth to form the Mangatepopo Stream (Fig. 2). Mangatepopo
Stream then flows northwards, before forming a tributary of the
Whanganui River, which is a major drainage channel for easterncentral
North Island.
At the head of the Mangatepopo valley, South Crater is a large
bedrock amphitheatre, which is 1 km across at its widest point and
250 m deep (Fig. 2; Fig. 3a). At 1967 m asl, Mt. Tongariro forms the
highest point on the south-facing back wall of South Crater, while
the Holocene-aged volcanic cone of Mt. Ngauruhoe (2287 m asl) is
situated c. 3 km to the south. Following Mathews (1967), we
interpret South Crater as a glacial cirque that served as the main
accumulation centre for the former Mangatepopo glacier. The flatbottomed,
cirque floor is c. 1700 m asl, although this contemporary
elevation represents infilling by an unknown thickness of postglacial
volcanic products. Bedrock spurs forming the mouth of the
cirque have been partially overprinted by post-glacial lavas sourced
from Mt. Ngauruhoe, however these spurs clearly curve towards
the southwest, which indicates the direction of ice flow from this
former accumulation zone.
To the southwest of Mt. Tongariro peak, the outer, westernfacing
slopes of South Crater exhibit a stacked sequence of truncated
lava flows in a steep, concave amphitheatre, which is suggestive
of headward erosion by a westward-flowing glacier. Field
investigations show that bedrock exposures in this region exhibit
glacially-polished surfaces, indicative of glacial abrasion by a
temperate ice-mass (Fig. 2). Several radiometric (K/Ar) ages from
lavas show that cone-building in the South Crater region occurred c.
70e100 ka (Hobden et al., 1996). These dates provide a maximum
bracketing age for the creation of the erosional glacial landforms.
The surface of the main valley floor is convex, representing postglacial
infilling by lava flows and pyroclastics from the nearby
volcanic vents. Both valley sides exhibit truncated lava cliffs,
aligned parallel to the west-east trending valley axis, and are
interpreted as having been eroded by ice. On the northern edge of
Pukekaikiore (Fig. 2A), two such vertical cliffs are stacked on top of
one another, separated by a clear bench, perhaps indicative of
multiple erosive periods by ice masses of differing thickness. K/Ar
dates from Pukekaikiore lavas range between c. 200e100 ka
(Hobden et al., 1996), which suggests that the glacial erosion
occurred during the last glacial cycle.
Multiple lateral moraines exist on both valley sides and can be
differentiated chronologically, based on morphostratigraphic
criteria. This distinction is clearest on the lower portion of the
northern valley side where two large lateral moraines have
significantly different crest heights and widths. Topographic profiles,
aligned perpendicular to the moraine crest orientation illustrate
the differing morphologies of these landforms (Fig. 2B). The
outer moraine (M3) is c. 20 m tall, with a highly rounded crest,
while the inner moraine (M2) is sharp-crested and significantly
taller. We interpret these geomorphic differences as representing a
considerable time gap between the two moraine forming glacial
episodes, with the outer, older moraine (M3) having undergone a
greater amount of degradation.
The lower portions (< 1200 m asl) of both M2 and M3 moraines
have continuous coverage by palaeosols, interbedded with tephra
and volcanic loess accumulation (Fig. 2; Fig. 3d,e). Between 1200
and 1250 m asl, these coverbeds become discontinuous, with the
transitions from bare moraine to the coverbeds marked by
erosional scarps, which indicate former, more extensive coverage of
the moraine surface at this elevation. The upstream end of moraine
M2 (>1250 m asl) is characterised by continuous thin (<1 m) coverbeds
on the lower moraine slope, which thins towards the bare,
rocky moraine crest (Fig. 3e). No erosional scarps exist on the upper
portion of this moraine.
Directly upstream from moraine M2, moraine M1 is a single,
continuous ridge (Fig. 3c), separated from M2 by a cliff, which
represents the downstream limit of an underlying lava flow dated
by Hobden et al. (1996) to 110 ± 6 ka. There is no tephra cover on
M1 present today, nor are there any remnants to suggest that
coverbeds have been more extensive in the past. It is likely that the
greater elevation of M1 prevents any accumulation of fine-grained
sediment due to wind exposure and lack of vegetation.
On the southern valley side, a single, sharp-crested lateral
moraine extends westwards for c. 1 km, from Pukekaikiore (Fig. 2).
The crest of this moraine is c. 40 m above the present day valley
floor. Immediately down valley, two left latero-frontal moraines
further constrain the former glacier margin. Soil and tephra cover
S.R. Eaves et al. / Quaternary Science Reviews 132 (2016) 40e5642
on these moraines precludes sampling for surface exposure dating,
although both landforms exhibit relatively sharp crests, which
suggests that they correlate with the M2 moraine on th eopposite
valley side. The innermost moraines on both valley sides converge
downstream at c. 1150 m asl, forming the valley mouth, thereby
defining the maximum limit of the most recent period of valley
glaciation (Fig. 3d).
2.2.2. Makahikatoa Stream
To the south of Mangatepopo valley, the northern headwaters of
the Makahikatoa stream drain southwestwards from a northeastsouthwest
orientated bedrock amphitheatre, interpreted as a
Fig. 2. (A) Glacial geomorphology and chronology of western Tongariro massif. Yellow stars represent the location of moraine boulder samples for cosmogenic 3
He exposure dating.
Associated labels are the calculated exposure ages (an internal unceratinties), using ’Lm’ scaling e see text for detailed description of age calculation methods. (B) Topographic
profiles across moraines M2 and M3, with 3
He exposure ages and sample numbers. Note the greater roundness of the older, M3 crest. *
Exposure ages for M3 are considered
minimum ages for moraine formation e see text.
S.R. Eaves et al. / Quaternary Science Reviews 132 (2016) 40e56 43
glacially eroded cliffs cut into the south-east flank of Pukekaikiore
(Fig. 2). The eastern limits of this former glacial catchment are illdefined
owing to the Holocene lava flows of Mt. Nguaruhoe,
which is situated immediately to the east.
A sharp-crested lateral moraine extends from the southern flank
of Pukekaikiore, descending in a southwest direction for c. 500 m,
terminating at c. 1275 m asl (Fig. 2). This moraine is paired by a
shorter moraine ridge, issuing from a steep bedrock spur that defines
the southern margin of this former glacial catchment. The
right lateral moraine cross-cuts a significantly wider, more rounded
moraine ridge that can be traced down valley to c. 1200 m asl and
represents a significantly older glacial limit. Based on the
morphostratigraphic relationship of these moraines, and their
similarity in form to the moraines in Mangatepopo valley, we
correlate them to the outer (M3) and inner (M1/M2) moraine ridges
in Mangatepopo valley, respectively (Fig. 2).
3. Methods
3.1. Cosmogenic surface exposure dating
Moraine boulder samples were collected using a portable rock
saw fitted with a segmented, diamond-tipped blade (e.g. Suganuma
et al., 2012). Samples were only taken from boulders on the
Fig. 3. (a) South Crater cirque, viewed from the upper slopes of Mt. Ngauruhoe; (b) moraines M2 (middle ground, left) and M3 (middle ground, centre), note the difference in
morphometry between the two landforms; (c) a view down valley from the upstream end of moraine M1 e note the glacially truncated side of Pukekaikiore in shadow on the far
valley side; (d) the inner lateral moraines of Mangatepopo valley clearly depict the former glacier terminus (photo taken from the upper western slopes of Pukekaikiore); (e)
coverbeds on the ice-proximal flank of M2 moraine thin towards the crest; (f) the sharp-crested right lateral moraine of the Makahikatoa valley runs diagonally from left to right
across this image (photo taken from the upper northern slopes of Pukekaikiore).
S.R. Eaves et al. / Quaternary Science Reviews 132 (2016) 40e5644
moraine crest (Fig. 4), and those close to known faults and beneath
lava cliffs were avoided. Where possible, we sampled boulders that
were partially embedded in the moraine matrix to minimise the
likelihood that the boulder has undergone post-depositional rotation.
All samples were collected from the highest point of the
parent boulder, and all boulders were over 0.6 m tall, thus minimising
the potential for past burial by snow and/or volcanic ash.
Sampled boulder surfaces exhibited <12

dip relative to horizontal,
thereby avoiding self-shielding. Azimuthal inclinations were
measured in the field using a compass and clinometer and geometric
shielding corrections were computed using the CRONUSEARTH
calculator (available at: http://hess.ess.washington.edu).
All shielding corrections were less than 1% (Table 1). Sample locations
and elevations were recorded using a Trimble GeoXH global
positioning system, relative to the WGS84 datum. These data were
differentially corrected using continuous measurements from
GeoNet ‘Chateau Observatory’ (‘VGOB’) base station (39.20

S,
175.54

E; 1161 m asl), located 8 km south west of Mangatepopo
Fig. 4. Examples of boulders sampled for cosmogenic 3
He surface exposure dating. Visible sample bags denote sample position on boulder surface. (A) Sample MP1201 at the
eastern end of moraine M1. Photo facing westwards. (B) Sample MP1203 on moraine M1 in the foreground, with Mt. Tongariro (1963 m asl) in the background. Photo facing
northeast. (C) Fracturing on sample MP1203. (D) MP1206 on moraine M2. Note prominent lateral moraines on opposite valley side. Photo facing southwest. (E) MP1207 on moraine
M2. Photo facing east. (F) Sample MP1208 on moraine M3. Note the highly diffuse moraine crest, compared to moraine M1. Also, note the Holocene cone of Mt. Ngauruhoe
(background right). Photo facing east. (G) Weathered surface of sample MP1209. (H) Samples MP1209 and MP1210 on moraine M3. Photo facing east.
S.R. Eaves et al. / Quaternary Science Reviews 132 (2016) 40e56 45
valley. Horizontal and vertical post-processed uncertainties for
individual sample locations are <1 m.
Whole rock samples were jaw-crushed, rinsed in de-ionised
water, and then dry-sieved to isolate the 250e500 mm size fraction.
Density (>3.1 g cmÀ3
) and magnetic techniques were used to
separate 150e600 mg of clinopyroxene (pigeonite). Crushing and
mineral separation were undertaken using facilities at Victoria
University of Wellington. Pyroxene separates were then prepared
according to Bromley et al. (2014) at the Cosmogenic Nuclide Laboratory
of Lamont-Doherty Earth Observatory (LDEO). Mineral
separates were first leached in 5% hydrofluoric (HF)/2% nitric
(HNO3) acid solution for 24 h to remove adhering ground mass
particles, followed by a separate 10% hydrochloric (HCl) acid solution
for 24 h. Leached pyroxenes were visually inspected for purity
and wrapped in aluminium foil.
Each sample was completely degassed by heating in a furnace to
>1300

C for 15 min, during which, released gases were exposed to
a liquid-nitrogen chilled, charcoal trap. Extracted gases were
exposed to an SAES getter before being collected on a cryogenic
cold trap at <15 K. Exclusively helium was then released by heating
the cold trap to 45 K. Mass spectrometry was conducted at the
LDEO Noble Gas Mass Spectrometry Laboratory using a MAP
215e50 noble gas mass spectrometer. Measurements were made
relative to the Yellowstone ‘Murdering Mudpot’ (MM) helium
standard (3
He/4
He ratio of 16.45Ra, where
Ra ¼ 3
He/4
Heair ¼ 1.384 Â 10À6
), using the protocol of Winckler
et al. (2005).
Attenuation of cosmogenic neutron flux with depth from the
surface was corrected for using measured sample thickness, a
standard rock density of 2.7 g cmÀ3
and an attenuation length of
160 g cmÀ2
(Dunne et al., 1999). Field observations and inspection
of satellite imagery indicate that the sampled boulders are not
subject to prolonged burial by seasonal snow. Snow cover is not
quantifiable for the geological past, however the topographic
prominence of the boulders and moraine crests is not conducive to
snow accumulation due to preferential erosion of snow by the
wind. Thus, we do not apply any correction to the age calculation
for burial by snow. The absence of resistant mineral veins in the
local andesites precluded quantitative assessment of postdepositional
erosion of boulder surfaces. Although sampled boulders
did not exhibit glacial striae, some boulders retain faceted
sides and care was also taken to avoid boulders that displayed clear
evidence of erosion, such as discolouration, onion-skin weathering,
and water pooling. We therefore choose to present the exposures
ages without an erosion correction.
Regional uplift in the central North Island has been estimated at
c. 0.6e1.0 mm yrÀ1
for the past 500 ka (Pulford, 2002). Integrating
this uplift rate into the age calculations for boulders exposed during
the LGM (c. 20e30 ka) does not alter the results outside the range of
the 3
He measurement uncertainty. For the oldest samples of this
study, integrating this uplift rate has the effect of increasing the
exposure ages by 1e2 kyr. We present the age dataset without
corrections for this effect, but discuss the implications for the
glacial chronology in the text.
We present surface exposure ages calculated using the CRONUScalc
MATLAB code of Marrero et al. (2015) (version 2.0) and the
Lal (1991)/Stone (2000)/Nishiizumi et al. (1989) (‘Lm’; Balco et al.,
2008) scaling model. Recently, Lifton et al. (2014) and Borchers
et al. (2015) have shown that this scaling model outperforms
neutron monitor based schemes (e.g. Dunai, 2001; Desilets et al.,
2006; Lifton et al., 2008). Calculation of surface exposure ages
from cosmogenic 3
He concentrations requires accurate constraint
of the local production rate. We take advantage of the recent,
proximal calibration of cosmogenic 3
He production site situated
<10 km from Mangatepopo valley (Eaves et al., 2015), which shows
that the local cosmogenic 3
He production rate agrees well with
global compilations of 3
He calibration datasets (e.g. Goehring et al.,
2010; Borchers et al., 2015). We use the compiled production rate
calibration of Borchers et al. (2015) implemented within the CRONUScalc
code.
3.2. Tephrochronology
Use of discrete, isochronous, pyroclastic (primarily ash and
lapilli) marker horizons as chronological tie-points in sedimentary
archives of paleoenvironmental change is well established in New
Zealand (e.g. Lowe et al., 2008, 2013). In glaciated landscapes, welldated
tephra layers stratigraphically overlying depositional landforms
such as moraines, provide minimum ages for moraine formation,
whereas tephra preserved beneath deposits of glacial
outwash or till, constrains the maximum age of deposition (e.g.
Kirkbride and Dugmore, 2001).
Since the work of Topping and Kohn (1973), significant progress
has been made to constrain the precise age and geochemical
signature of rhyolitic tephra marker horizons in New Zealand (e.g.
Lowe et al., 2008, 2013; Vandergoes et al., 2013). To test the original
interpretation of Topping and Kohn (1973), we revisited the soil
section overlying moraine M2 in Mangatepopo valley to sample the
lower two rhyolitic tephra horizons for electron microprobe (EMP)
analysis of glass shard major element compositions. To aid identification,
we compare the geochemistry of the sampled tephras to
that of a discrete rhyolitic tephra horizon identified at Taurewa road
cutting (39.12

S, 175.52

E; 825 m asl), situated c. 5 km west from
the study site (Fig. 1). This tephra stratigraphically overlies the
Kawakawa/Oruanui Tephra (KOT), which was recently dated at this
section (c. 25.4 cal ka; Vandergoes et al., 2013).
Table 1
Location and geometry of cosmogenic3
He samples.
Sample ID Lat. (
S) Long. (
E) Altitude (m asl) Thickness (cm) Surface strike/dip Shielding
Moraine M1:
MP1201 39.136 175.619 1519 2.0 0 0.998
MP1202 39.138 175.612 1429 2.0 210/10 0.998
MP1203 39.138 175.612 1429 2.0 004/4 0.998
MP1204 39.137 175.618 1481 1.5 0 0.997
MP1205 39.138 175.614 1437 2.0 312/8 0.997
Moraine M2:
MP1206 39.141 175.598 1302 2.5 098/10 0.997
MP1207 39.141 175.600 1308 2.5 0 0.998
Moraine M3:
MP1208 39.139 175.598 1286 2.5 0 0.998
MP1209 39.139 175.598 1280 3.0 258/6 0.998
MP1210 39.139 175.598 1275 2.5 296/2 0.998
MP1211 39.139 175.597 1266 3.0 124/2 0.998
S.R. Eaves et al. / Quaternary Science Reviews 132 (2016) 40e5646
All tephra samples were wet-sieved to isolate the >63 mm
fraction, before individual glass shards were handpicked (n ¼ >18
per sample; Table 3) and loaded in an epoxy mount. All major
element determinations were made on a JEOL Superprobe (JXA-
8230) at Victoria University of Wellington, using the ZAF correction
method. Analyses were performed using an accelerating voltage of
15 kV under a static electron beam operating at 8 nA. The electron
beam was focused to 10 mm.
3.3. Equilibrium line altitude reconstruction
Palaeoclimatic calculations from glacier reconstructions rely on
accurate delineations of former ice masses, which are constrained
by interpretation of ice-marginal geomorphology (Porter, 1975;
Benn et al., 2005). In Mangatepopo valley, moraine ridges and
glacial till extent clearly depict the terminus and lower margins of
the former glacier, while erosional landforms such as glacial cirque
and glacially-trimmed cliffs provide guidance on the upper glacier
limits. We reconstruct the palaeo-equilibrium line altitude (pELA)
of the Mangatepopo glacier for the LGM only, as the older ice limits
are less well preserved. In the vicinity of Mt. Ngauruhoe, postglacial
volcanic activity has obscured evidence of the former
glacial limits, therefore we estimate former ice limits in this region
by interpolating between the cirque headwall and the glaciallytruncated
lava flows of Pukekaikiore (Fig. 2). To estimate the
hypsometry of the former glacier, elevation contours are drawn
from the intersection of the reconstructed ice margin with the
modern topographic contours. Ice surface topography is reconstructed
to represent flow vectors in contemporary glaciers, which
are increasingly convergent with distance upstream from the ELA,
and increasingly divergent downstream (Paterson, 1994). We
define the upper margins of the LGM glacier using the head of
South Crater cirque (sensu Kaplan et al., 2010), although, it is
possible that additional ice was sourced from an ice field centred on
Tongariro massif (Section 2.2.2).
A number of methods exist to estimate the pELA for former
glaciers. The Accumulation Area Ratio (AAR) method is applied
most commonly, which is based on the assumption that the accumulation
zone of a steady-state glacier represents a fixed proportion
of the total glacier area. Accumulation areas of modern glaciers
globally, typically occupy 50e80% of the total glacier surface area
(Meier and Post, 1962). Empirical studies show that New Zealand
Table 3
Glass shard major element compositions of rhyolitic tephras from the Taurewa and Mangatepopo sections, compared with potential correlatives from the Okataina Volcanic
Centre (OVC). Oxide values are recalculated to 100% on a volatile-free basis. Total Fe expressed as FeOt. Mean and 1 standard deviation (italics), based on n analyses. All samples
normalised either against glass standard VG-568 or ATHO-G. EMP Analyst: B.V. Alloway, for all samples except ‘MP833d’ (S.R. Eaves).
SiO2 Al2 O3 TiO2 FeO MgO MnO CaO Na2 O K2 O Cl Total n
Mangatepopo moraine:
MP833(i) 78.13 12.39 0.15 0.98 0.13 0.05 0.89 3.96 3.23 0.09 98.21 20
0.28 0.12 0.03 0.08 0.01 0.03 0.04 0.14 0.10 0.01 1.28
MP833a(ii) 78.21 12.38 0.14 1.00 0.12 0.05 0.88 3.84 3.28 0.09 98.2 20
0.29 0.1 0.03 0.14 0.03 0.03 0.04 0.14 0.31 0.02 1.58
MP833d 78.86 12.39 0.14 0.91 0.13 0.05 0.88 3.5 2.99 0.15 99.3 18
0.24 0.10 0.02 0.09 0.03 0.03 0.06 0.14 0.12 0.02 1.00
Taurewa (SHW-47):
Waiohau (13-16-27) 78.23 12.31 0.14 0.96 0.11 0.05 0.87 3.96 3.27 0.09 98.48 21
0.25 0.12 0.02 0.07 0.02 0.03 0.05 0.23 0.16 0.01 1.66
Reworked KOT? (13-16-29) 78.14 12.39 0.16 1.1 0.14 0.04 0.99 3.74 3.2 0.1 97.21 20
0.22 0.07 0.03 0.09 0.04 0.03 0.03 0.12 0.17 0.02 1.55
Oruanui (KOT) (13-16-30) 78.06 12.41 0.12 1.19 0.11 0.04 1.02 3.75 3.19 0.11 96.42 22
0.19 0.10 0.02 0.07 0.02 0.03 0.06 0.14 0.10 0.01 0.82
Standard - VG-568:
Sept. 14 2012 75.6 12.2 0.23 3.27 0.1 0.1 1.72 3.75 2.61 nd 99.57 31
0.55 0.12 0.02 0.09 0.01 0.02 0.03 0.1 0.03 0.65
Standard - ATHO-G:
Oct. 1 2013 75.57 12.2 0.26 3.27 0.09 0.1 1.7 3.73 2.64 0.02 99.58 48
0.60 0.09 0.03 0.10 0.01 0.03 0.03 0.13 0.05 0.01 0.75
Dec. 2013 75.62 12.2 0.24 3.27 0.09 0.11 1.7 3.73 2.64 0.08 99.68 20
0.36 0.07 0.02 0.09 0.01 0.04 0.02 0.11 0.06 0.06 0.41
Table 2
Helium measurement data and exposure ages for all samples.3
Hec concentrations are derived using an assumed pyroxene crystallisation age of 110 ka (see Section 4.1). Internal
uncertainty of exposure ages includes the 1s analytical error and an assumed 20% error from the magmatic3
He correction. Exposure ages are calculated using the scaling model
of Lal (1991)/Stone (2000)/Nishiizumi et al. (1989), labelled ‘Lm’ by Balco et al. (2008).
SampleID 3
Hetotal ± 1s (106
at. gÀ1
) R/Ra
3
He ± 1s (106
at. gÀ1
) D3
He (%) Exposure age ± 1s int. (Ext.) (ka)
Moraine M1:
MP1201 11.8 ± 0.28 156 11.5 ± 0.28 À2.3 30.0 ± 0.7 (3.2)
MP1202 8.28 ± 0.17 138 8.10 ± 0.18 À2.3 23.0 ± 0.5 (2.5)
MP1203 5.64 ± 0.18 73 5.34 ± 0.19 À5.3 15.5 ± 0.5 (1.7)
MP1204 8.39 ± 0.34 172 8.27 ± 0.34 À1.4 22.5 ± 0.9 (2.5)
MP1205 9.23 ± 0.20 179 9.10 ± 0.21 À1.5 25.6 ± 0.5 (2.7)
Moraine M2:
MP1206 6.34 ± 0.17 47 5.70 ± 0.22 À10.1 18.2 ± 0.7 (2.0)
MP1207 6.86 ± 0.19 80 6.52 ± 0.20 À5.0 20.5 ± 0.6 (2.1)
Moraine M3:
MP1208 18.9 ± 0.33 302 18.7 ± 0.33 À1.1 57.4 ± 1.0 (6.4)
MP1209 17.4 ± 0.31 190 17.0 ± 0.32 À2.2 52.5 ± 1.0 (6.1)
MP1210 15.2 ± 0.33 195 14.9 ± 0.33 À1.9 45.9 ± 1.0 (5.1)
MP1211 16.9 ± 0.32 227 16.6 ± 0.33 À1.6 51.8 ± 1.1 (6.0)
S.R. Eaves et al. / Quaternary Science Reviews 132 (2016) 40e56 47
glaciers most commonly have an accumulation-ablation area ratio
of 2:1 (AAR ¼ 0.67; Chinn et al., 2012).
A shortcoming of the AAR method is the failure to account for
glacier hypsometry (Furbish and Andrews, 1984). Reconstructions
of former valley and cirque glaciers are usually well-constrained by
topography, therefore AAR-based pELA estimates are subject to
lower uncertainty. However, the hypsometry of former valley glaciers
sourced from ice caps or plateau ice fields is less certain due to
the absence of geomorphic indicators that constrain ice thickness in
the upper catchment. Underestimating the planimetric surface area
at the glacier head leads to underestimation of the pELA, as a larger
ablation area is required to balance the expanded accumulation
zone. Other techniques have been developed to overcome this
shortcoming in the AAR method (Furbish and Andrews, 1984;
Osmaston, 2005), however these are also reliant on accurate
knowledge of the former ice geometry. The glacial geomorphology
of the wider Tongariro massif suggests that the former Mangatepopo
glacier may have been sourced from a central ice field, as
several other moraine-lined glacial valleys radiate from the centre
of the edifice (Mathews, 1967), thus implying a central source. If
this were the case, the AAR method will underestimate the pELA,
therefore we consider the AAR-based reconstructions as a
maximum estimate of ELA depression from present.
Reconstructed pELAs provide a useful metric for comparing the
magnitude of past climate change (Porter, 1975). This requires accurate
knowledge of the present-day ELA. No glaciers currently
exist on Tongariro massif, however small cirque glaciers persist on
Mt. Ruapehu (2797 m asl), situated 15 km to the south. In the most
recent survey of these glaciers, Keys (1988) reports contemporary
ELAs of 2340e2650 m asl, based on end of summer snowline surveys.
Topoclimatic factors such as wind-driven snow accumulation
and topographic shading impart a strong influence on the mass
balance of cirque glaciers (Kuhn, 1995), which can reduce the utility
of the ELA as an index for atmospheric temperature change.
However, Brook et al. (2011) analysed surface area changes of one
glacier on Mt. Ruapehu between AD1988e2008 and found that
variations closely follow ablation season temperature changes. This
correlation supports the use of end of summer snowline observations
on Mt. Ruapehu to relate our pELA reconstructions to past
atmospheric temperature change. We take the arithmetic mean of
the midpoints of Keys (1988) observations (2483 ± 55 m asl) to use
as a modern (AD1988) ELA datum.
4. Results
4.1. Cosmogenic 3
He concentrations
Total measured 3
He (3
Hetotal) in a sample can comprise cosmogenic
(3
Hec), magmatic (3
Hem), and nucleogenic (3
Hen) components.
We derive 3
Hec from our measurements of 3
Hetotal, using the
following equation:
3
Hec ¼ 3
Hetotal À 3
Hen À 4
Hem  ð3
He=4
HeÞm (1)
3Hen is a product of the mineral composition and crystallisation
age (Andrews, 1985). Our samples exhibit low U, Th and Li concentrations,
which limits 3Hen production (Supplementary
Table A1). Furthermore, the parent lavas of the moraine boulders
sampled in the Mangatepopo valley are relatively young (c.
60e160 ka; Hobden et al., 1996), therefore there has been limited
time for 3
Hen accumulation. The contribution of 3Hen to 3
Hetotal is
therefore negligible for our samples and we neglect this term from
Equation (1).
3
Hem is sourced from the mantle, where 3
He/4
He ratios are
enriched relative to the atmospheric helium ratio
(Ra ¼ 1.384 Â 10À6
). Measurements of magmatic helium in hightemperature
fluids and shielded clinopyroxene phenocrysts from
TgVC consistently yield 3
He/4
He ratios of c. 6Ra (Torgersen et al.,
1982; Patterson et al., 1994; Hulston and Lupton, 1996), therefore
we use this value for our calculations.
To calculate magmatic 4
He (4Hem), we use U and Th concentrations
from two samples (Supplementary Table A1) to derive the
in situ production rate of radiogenic 4
He (4
Her) according to
Andrews (1985), and subtract this radiogenic contribution from
4
Hetotal (4
Hem ¼ 4
Hetotal À 4
Her). Neglecting 4
Her can result in
overcorrection for 3
Hem, thus causing underestimation of 3
Hec
(Blard and Farley, 2008). U and Th concentrations are 0.019e
0.023 ppm and 0.141e0.174 ppm, respectively (Supplementary
Table A1), which yield in situ 4
Her production rates (P4) of
1.82e1.95 Â 105
at. gÀ1
yrÀ1
(Supplementary Table A2). Given the
similarity of the U and Th measurements between samples, we use
the average of the two sample concentrations for samples without
compositional data, which yields a 4
Her production rate of
1.89 Â 105
at. gÀ1
yrÀ1
.
To estimate 4
Her, we use a pyroxene crystallisation age of 110 ka,
which is the mid-range of the radiometric ages of Mangatepopo
valley lavas dated by Hobden et al. (1996) (c. 60e160 ka). Using this
age means that 4
Her accounts for 21e59% of 4
Hetotal in our samples.
Substituting this calculation into Equation (1) indicates that
90e99% (median ¼ 98%) of 3
Hetotal in our samples is of cosmogenic
origin (Table 2). Use of older crystallisation ages increases the 4
Her
component, and vice versa for younger crystallisation ages
(Supplementary Table A2). Altering the crystallisation age by
±50 ka changes the 3
Hec concentrations by an average of ±0.8%,
relative to the 110 ka calculation (Supplementary Table A2). Thus
our 3
Hec concentrations are relatively insensitive to the choice of
crystallisation age. We present exposure ages for all samples using
3
Hec reported in Table 2, which includes a 20% error for the
magmatic correction propagated into the 1s analytical uncertainty.
4.2. Cosmogenic 3
He results and moraine age interpretation
4.2.1. Upper moraine (M1)
Exposure ages (with internal uncertainties) of the five boulders
sampled from moraine M1 are 30.0 ± 0.7 ka, 25.6 ± 0.5 ka,
23.0 ± 0.5 ka, 22.5 ± 0.9 ka and 15.5 ± 0.5 ka (Table 2). We adjudge
the youngest sample (MP1203) to be an outlier, based on the
preservation of moraine M2 down valley, which dates to 18e20 ka
(see section 4.2.2 below). It is unlikely that M2 would be preserved
today, if ice of sufficient thickness to be depositing boulder MP1203
at the elevation of M1 was present in Mangatepopo valley at c.
16 ka. Re-examination of field descriptions and photographs (e.g.
Fig. 4B) suggests that this boulder may have shed a portion of its
surface since deposition, as evidence of fracturing is present,
probably caused by preferential weathering along internal cooling
joint planes. Post-depositional removal of the boulder surface
through this process provides a possible explanation for the
anomalously young age.
Following omission of this geomorphic outlier, three possible
depositional scenarios can explain the age distribution of the
remaining four samples. First, the oldest age (c. 30 ka e MP1201)
may represent inheritance of cosmogenic 3
He from exposure prior
to deposition at its current position on the moraine crest. In this
scenario, the true age of M1 would be closer to the remaining three
samples, between 26 and 23 ka. Second, the oldest age (c. 30 ka e
MP1201) may represent the true age of M1, with the remaining
samples having been subject to post-depositional erosion of the
boulder surface, or surface shielding (e.g. Ivy-Ochs et al., 2007),
causing the surface exposure ages to post-date moraine formation.
Third, M1 could represent a composite landform (e.g.
S.R. Eaves et al. / Quaternary Science Reviews 132 (2016) 40e5648
Roethlisberger and Schneebeli, 1979) comprising deposits from
several glacier fluctuations of similar magnitude between 30 and
23 ka.
Based on the existing dataset, it is not possible to unequivocally
attribute either one of these scenarios to the M1 moraine. However,
analyses of large datasets of moraine boulder surface exposure ages
show that only a very small percentage of boulders deposited by
non-polar, temperate glaciers exhibit cosmogenic nuclide inheritance
from periods of prior exposure. For example, recent dating
campaigns in the Southern Alps that have generated several hundred,
individual high-precision in situ cosmogenic 10
Be exposure
ages for moraine boulders (Schaefer et al., 2009; Kaplan et al., 2010;
Putnam et al., 2010a, 2013a, 2013b; Kelley et al., 2014), which
indicate that any inherited component is typically less than the
measurement uncertainty (see Schaefer et al., 2009 for a detailed
analysis). The maritime, mid-latitude location of our study site,
together with geomorphic evidence for scouring of the glacier bed
(Section 2.2.1) suggest that former glaciation at this site was warmbased.
Thus, while it is feasible that the 3
He content of MP1201
could represent inheritance from prior exposure, we consider this
scenario the least likely explanation for the age distribution.
It is more difficult to decipher between the latter two scenarios,
however both imply moraine aggradation at c. 30 ka, with the
remaining younger ages either representing post-depositional
disturbance (scenario 2), or further aggradation from glacial reoccupation
at 26e23 ka (scenario 3). While scenario 2 is commonly
assumed in such situations, diachronous moraines have also previously
been inferred from surface exposure age datasets displaying
multiple populations (e.g. Licciardi et al., 2004; Briner, 2009).
This is more likely in situations where a glacier is topographically
constrained (Roethlisberger and Schneebeli, 1979), such as the
Mangatepopo valley. Exposure ages from down-valley (Section
4.2.2) suggest valley glaciation endured until c. 18e20 ka, therefore
we suggest the age distribution from M1 most likely represents
time-transgressive aggradation from persistent valley glaciation
through the period c. 30e23 ka. Moraine construction would have
occurred when the ice surface elevation was sufficient to overcome
the prominent lava flow that bounds the northern valley side.
Furthermore, we note that this interpretation is consistent with
evidence from other glaciated catchments in New Zealand, which
indicate glacier termini were at or near their maximum downstream
limits for the last glacial cycle between c. 30e18 ka (e.g.
Putnam et al., 2013b; Kelley et al., 2014; Rother et al., 2014; see
Discussion).
4.2.2. Lower moraine e inner (M2)
Two boulders were deemed suitable for exposure dating
(MP1206 and MP1207; Fig. 4C and D) and returned ages of
18.2 ± 0.7 ka and 20.5 ± 0.6 ka, respectively. These ages are in close
agreement and are stratigraphically coherent with the age of M1
and the overlying tephrostratigraphy (Section 4.3).
4.2.3. Lower moraine e outer (M3)
Samples were collected from four boulders, which returned ages
ranging from 57.4 ± 1.0 ka to 45.9 ± 1.0 ka. We consider these ages
to be minimum-limiting ages for this landform for the following
reasons.
Continual exposure of these andesitic boulders to the relatively
high total annual precipitation at this location (c. 3 m yrÀ1
; NIWA,
2014), over several tens of millennia, is likely to have caused
granular disintegration of the boulder surfaces. Such erosion would
remove cosmogenic 3
He atoms, thus resulting in nuclide concentrations
that underestimate the true exposure age of the boulder.
For example, if the sampled boulders on moraine M3 have been
exposed since c. 63 ka, which corresponds with the culmination of a
major glacier advance in the Southern Alps (Schaefer et al., 2015),
then surface erosion rates of 1.5e5.5 mm kyrÀ1
are required to
replicate our observed exposure ages. Such rates of rock erosion
agree well the compilation of Portenga and Bierman (2011) which
indicates typical long term erosion rates of 1e10 mm kyrÀ1
for
igneous lithologies in temperate climatic environments.
Our age calculations are uncorrected for potential tectonic uplift
during the period of exposure, which has been estimated at c.
0.6e1.0 mm yrÀ1
(Pulford, 2002). Including c. 40e60 m of uplift into
the exposure age calculation serves to lower the time-integrated
local production rate relative to that estimated for the present
day sample elevation. Depending on the temporal pattern of uplift
over the exposure period, this effect could contribute to underestimation
of boulder ages by up to 2 kyr.
All boulders on the moraine have been subject to similar tectonic
uplift, thus this effect should be uniform between samples.
The effects of boulder surface erosion is likely to have impacted
individual boulders to varying degrees (e.g. due to differences in
rock hardness), thus contributing to the scatter amongst the
exposure ages. We therefore consider that the oldest samples,
which yield exposure dates of 52e57 ka, represent the closest
minimum age constraint for the true age of moraine M3. However
the true moraine age may be several millenia older still (see
Discussion).
4.3. Coverbed stratigraphy and tephra major element geochemistry
Stratigraphic logging of soil sections overlying the ice-proximal
flank of the M2 moraine (Fig. 5A) permits correlation with the
detailed descriptions of Topping and Kohn (1973) and Topping
(1974). At c. 450 cm depth we identify a matrix-supported diamicton,
predominantly consisting of boulders and cobbles in a silty
matrix, which represents the surface of the underlying lateral
moraine. A section containing approximately 90 cm of interbedded
cm-to dm-thick andesitic ash and lapilli beds, with two yellowwhite
rhyolitic tephra beds and weakly formed palaeosols, immediately
overlies the moraine surface. The lowermost rhyolitic
tephra (sample: ’MP833a’) is 0.2e1 cm thick and situated c.
10e20 cm above the moraine surface, immediately overlying a 3 cm
thick, coarse (med. sand), dark-grey andesitic tephra (Fig. 5C). The
lower rhyolitic tephra exhibits ductile fold structures (Fig. 5B),
which suggests post-depositional deformation, perhaps via frostheave.
Approximately 40 cm above MP833a, a discontinuous
rhyolitic tephra (sample: ‘MP833d’) varies in thickness from 0 to
3 cm (Fig. 5C). We identify a distinct cobble-to-fine gravel unit at
350 cm depth, which we correlate to the pebble unit identified by
Topping and Kohn (1973) (described fully in Topping, 1974),
therefore we are confident that these tephra samples correspond
with those analysed by Topping and Kohn (1973). A sharp erosional
contact separates this coarse unit from a massive, grey/brown
silteclay bed (300e50 cm depth), with abundant rhizomorphs and
two, discrete, interbedded pumiceous horizons. The lower pumice
bed (at c. 275 cm depth) is c. 10 cm thick and consists of yellow and
dark brown coloured clasts, up to 2 cm in diameter. The upper
pumice is c. 40e50 cm thick and immediately underlies the modern
soil horizon. This pumice is correlated to the Taupo ignimbrite
(AD232 ± 10 yr; Lowe et al., 2013), based on stratigraphic position,
large pumice clasts (up to 10 cm diameter), and abundant charred
twigs.
Shane (2000) summarised the major element chemistry of post-
25 ka rhyolitic tephras in North Island and found that OVC tephras
are characterised by higher SiO2 (c. 76e79 wt. %) and lower FeO (c.
1 wt. %), compared to those from the Taupo Volcanic Centre
(SiO2 ¼ c. 71e77 wt. %; FeO ¼ 1.5e3.5 wt. %). Using this information,
we can assign samples MP833a and MP833d (Table 3; Fig. 6) to the
S.R. Eaves et al. / Quaternary Science Reviews 132 (2016) 40e56 49
OVC with high confidence. This supports the original interpretation
of Topping and Kohn (1973), who assigned both to horizons in the
Mangatepopo section to OVC-sourced eruptions. However, the
major element compositions of MP833a and MP833d are also
indistinguishable from one another (Table 3; Fig. 6), which suggests
that the upper sample (MP833d) could represent a reworked
remnant of MP833a below. This inference is further supported by
the field observations of deformation structures within the lower
horizon (MP833a), but contradicts the initial interpretation of
Topping and Kohn (1973). The geochemical composition of the
Taurewa rhyolitic tephra (Sample: 13-16-27; Fig. 5) is indistinguishable
from the Mangatepopo data, which suggests that it
represents the same volcanic event (Fig. 6).
To further constrain deglaciation in the Mangatepopo valley, we
seek to determine which OVC-sourced event is represented by the
Mangatepopo tephra. To do this, we compare the glass shard major
element measurements to proximal and distal OVC reference data
from Honeycomb Trench (B.V. Alloway, unpub. data) and Waipaoa
river basin (Bilderback, 2012; Marden et al., 2014), respectively. The
Mangatepopo and Taurewa data are sufficiently different from the
Rotorua Tephra for us to rule this out as a possible correlative
(Fig. 6). Using binary plots of K2OeFeO, the Rerewhakaaitu Tephra
can be readily discriminated from the Waiohau Tephra by the
presence of high and low K2O populations in the former (Fig. 6B;
Shane et al., 2008). The Mangatepopo samples display a single K2O
population, which is consistent with the Waiohau Tephra in the
proximal reference datasets (Fig. 6). Furthermore, biotite has been
noted as a diagnostic component of the ferromagnesian mineral
assemblage of the Rerewhakaaitu Tephra (Froggatt and Lowe, 1990)
and we do not identify any biotite flakes within the mineral assemblages
of the samples. Isopach maps of tephra dispersal over
North Island (Lowe et al., 2013) indicate a similar thickness
(0.5e1 cm) for both the Rerewhakaaitu and Waiohau tephras in the
vicinity of the study site, therefore field observations of thickness
are not useful in discriminating between the two.
In summary, field descriptions and EMP measurements of glass
Fig. 5. (A) Stratigraphic log of the Taurewa reference section and the Mangatepopo moraine section originally described by Topping and Kohn (1973), with the samples labelled; (B)
The lowermost rhyolitic tephra (sample MP833a) exhibiting deformation structures likely caused by gelifluction in a formerly frozen soil (Photo: B.V. Alloway); (C) Soil pit showing
both rhyolite horizons at the Mangatepopo section.
S.R. Eaves et al. / Quaternary Science Reviews 132 (2016) 40e5650
shard major elements suggest that the samples from two rhyolitic
tephra horizons likely represent the same event, as opposed to the
previous interpretation of two separate events (Topping and Kohn,
1973). Furthermore, the major element composition of this tephra
can be correlated to the OVC with high confidence. The body of
field-evidence, mineralogy and glass geochemistry all strongly
indicate that the ash layer is likely to be Waiohau Tephra, which
was emplaced at c.14.0 ± 0.2 cal ka BP (Lowe et al., 2013). This result
is stratigraphically consistent with surface exposure ages from the
moraine crest (c. 18e20 ka; Section 4.2.2) and accords with
moraine-tephrostratigraphy elsewhere on Tongariro massif (Cronin
and Neall, 1997) that suggests moraine abandonment and soil
aggradation was well underway prior to c. 14 ka, in response to
climatic amelioration. This represents a minimum age for iceretreat,
which likely occurred several thousands of years earlier,
as indicated by the cosmogenic surface exposure ages on the
moraine crest.
4.4. Equilibrium line altitude reconstruction
We derive a pELA (AAR ¼ 0.67; Chinn et al., 2012) for the LGM
Mangatepopo glacier of c. 1410 m asl (Fig. 7a). This is in reasonable
agreement with the maximum elevation of lateral moraine (MELM;
Andrews, 1975) M1 at c. 1530 m asl, which can be used to
approximate the pELA associated with the older (c. 30e23 ka),
slightly thicker Mangatepopo glacier. The moraine stratigraphy
suggests that the Mangatepopo glacier was broadly of similar
extent during the period 30e18 ka, therefore the MELM method
provides a useful independent test of the AAR-based ELA estimate,
particularly as it is not subject to uncertainties in past glacier
hypsometry or topographic change. Preservation of the slightly
older moraine M1 implies the pELA associated with M2 cannot be
lower than 1530 m asl. The lower value returned by the AAR
method is thus likely to reflect minor uncertainties in the palaeoglacier
reconstruction either caused by post-glacial topographic
change, and/or the former presence of a wider accumulation zone
than currently appreciated, such as an ice field (e.g. Rea et al.,1999).
Given the uncertainties in both methods of pELA reconstruction, we
consider 1400e1550 m asl a most-likely estimate of the ELA in
Mangatepopo valley during the period 30e18 ka. This represents an
ELA lowering of c. 930e1080 m, relative to the contemporary ELA
datum of Keys (1988) (2483 ± 55 m asl).
The ELA on a given glacier is primarily controlled by summer air
temperature and winter precipitation, although a range of other
energy-balance (insolation, local wind speed, cloudiness, humidity)
and topoclimatic (avalanching, snow drifting, topographic shading)
factors also contribute (Oerlemans and Fortuin, 1992). Assuming no
change in precipitation, it is possible to derive a first-order estimate
of atmospheric temperature change associated with a pELA
reconstruction, using a temperature lapse rate. Temperature lapse
rates can vary significantly in space and time (Doughty et al., 2013),
and ELA-based palaeotemperature reconstructions are sensitive to
this value. For example, we calculate an LGM temperature lowering
of c. 5.4

C relative to present, when using the mean annual temperature
lapse rate for upland (>300 m) New Zealand (À5.1

C
kmÀ1
; Norton, 1985). However, using the standard environmental
lapse rate (À6.5

C kmÀ1
) increases the temperature depression
estimate to 7.0

C. Fig. 7b shows the cumulative probability distribution
of palaeotemperature estimates in Mangatepopo valley
derived using the empirically constrained palaeo- (1400e1550 m
asl; AAR and MELM) and contemporary (2483 ± 55 m asl; Keys,
1988) ELAs, and a range of possible atmospheric temperature
Mangatepopo (MP-833a; n=40)
Mangatepopo (MP-833d; n=18)
This study
0.5 0.6 0.7 0.8 0.9 1.0 1.1 1.2 1.3 1.4 1.5
FeO (wt %)
2
3
4
5
K2
O
(wt %)
CaO
(wt %)
Waoihau Tephra, Honeycomb Trench (n=58)
Rerewhakaaitu Tephra, Honeycomb Trench (n=28)
Rotorua Tephra, Honeycomb Trench (n=50)
Proximal Okataina VC tephra
B.V. Alloway, unpublished data
(acquired on VuW electron microprobe)
Waoihau Tephra (n=26)
Rerewhakaaitu Tephra (n=32)
Rotorua Tephra (n=13)
0.5 0.6 0.7 0.8 0.9 1.0 1.1 1.2 1.3 1.4 1.5
2
3
4
5
FeO (wt %)
0.5 0.6 0.7 0.8 0.9 1.0 1.1 1.2 1.3 1.4 1.5
2
3
4
5
FeO(wt %)
Bilderback (2012)
(acquired on VuW electron microprobe)
0.5
0.6
0.7
0.8
0.9
1.0
1.1
1.2
1.3
1.4
1.5
0.5
0.6
0.7
0.8
0.9
1.0
1.1
1.2
1.3
1.4
1.5
0.5
0.6
0.7
0.8
0.9
1.0
1.1
1.2
1.3
1.4
1.5
Taurewa, SHW-47
(13-16-27) n=21
Fig. 6. Glass shard FeO vs. K2O and CaO (wt. %) plots of Mangatepopo samples compared with OVC-proximal (Honeycomb Trench) and north-eastern North Island distal (Waipaoa)
reference data. All major element EMP glass data was acquired on the same analytical instrument at Victoria University of Wellington under the same operational conditions and
using the same internal standards.
S.R. Eaves et al. / Quaternary Science Reviews 132 (2016) 40e56 51
lapse rates, equally weighted between 4 and 7

C kmÀ1
. This yields a
normally distributed range of palaeotemperature estimates, centred
on 5.6 ± 1.1

C (Fig. 7a), which we consider to be a bestestimate
estimate of temperature anomaly in central North Island
between 30 and 18 ka.
5. Discussion
5.1. The Last Glacial Maximum in central North Island
Using cosmogenic 3
He surface exposure dating of moraine
boulders, we provide the first direct chronological constraint for
extensive valley glaciation on Tongariro massif during MIS 3e2,
which began as early as 30 ± 3 ka and persisted until at least c.
18e20 ka. This is well-aligned, within dating errors, with the Last
Glacial Cold Period (LGCP; c. 29e18 ka) as identified in the New
Zealand Climate Event Stratigraphy (Barrell et al., 2013). The local
ELA depression associated with LGCP glaciation in Mangatepopo
valley was c. 1400e1550 m asl (Fig. 7). The timing and magnitude of
these changes are in good agreement with previous palaeoenvironmental
reconstructions from central North Island. On Mt.
Ruapehu, situated c. 15 km to the south of Mangatepopo valley,
McArthur and Shepherd (1990) report geomorphological evidence
for a former ice mass with an equilibrium line of 1500e1600 m asl,
which agrees well with our proximal pELA reconstruction.
McArthur and Shepherd (1990) suggest deformed pro-glacial lake
sediments interbedded with moraines on northeast Ruapehu record
multiple glacier advances during the last glacial cycle. Topping
(1974) identified the Kawakawa-Oruanui Tephra within these lake
sediments, which places glacier advances either side of 25.4 ka. The
latter advance likely correlates to the final stand of the Mangatepopo
glacier at 18e20 ka, however the lack of direct dating on
Mt. Ruapehu means the timing of preceding glacier fluctuations is
unconstrained.
Our new glacial chronology for central North Island exhibits
good agreement with the only other LGM glacial reconstruction in
North Island, New Zealand. Brook et al. (2008) report cosmogenic
10
Be moraine and bedrock exposure ages of c. 18 ka in Park Valley of
the Tararua ranges of southern North Island (c. 40

S). Recalculating
these ages using the local 10
Be production rate of Putnam et al.
(2010b) yields revised ages of c. 19e21 ka, which is indistinguishable
from the M2 moraine in the Mangatepopo valley. Brook (2009)
also identifies the KOT within the Park Valley moraine, which
suggests it represents a composite feature first occupied prior to
25.4 ka (Brook and Crow, 2008; Brook, 2009). This evidence is also
in agreement with our findings that the M1 and M2 moraines were
constructed between 30 and 18 ka.
Recent, high-precision cosmogenic 10
Be moraine chronologies
from the central Southern Alps show a similar temporal pattern of
glacier fluctuations. Local constraint of the 10
Be production rate
(Putnam et al., 2010b), coupled with favourable topographic situations
for moraine preservation and extensive exposure age datasets,
have afforded detailed insight to late Quaternary glacier
fluctuations in this region. For example, Putnam et al. (2013b) show
that the former Ohau glacier deposited terminal moraines at c.
32 ka, c. 27 ka c. 23 ka and c. 18 ka. Similarly, the nearby Pukaki
glacier attained its maximum LGCP length by 28 ka and fluctuated
about this position until 18 ka (Schaefer et al., 2006; Kelley et al.,
2014; Schaefer et al., 2015). On the west coast of South Island,
Suggate and Almond (2005) suggest glacier advances culminated at
c. 28 ka, 22 ka and 19 ka, however recent cosmogenic 10
Be surface
exposure dating of these moraine sequences has refined the age of
these deposits to c. 25 ka, c. 21 ka and c. 17 ka (Barrows et al., 2013).
On the eastern side of the Southern Alps, Rother et al. (2014) show
that the former Rangitata glacier reached its maximum extent
before c. 28 ka, followed by successive fluctuations of slightly lesser
extent between 26 and 19 ka. In the south, Williams et al. (2015)
report that speleothem growth in Aurora Cave ceased at c. 33 ka
when a glacier advanced in the Te Anau trough. Thus, our findings
from central North Island add to a growing body of evidence that
show glaciers across New Zealand attained their maximum extent
in late MIS 3 and fluctuated about this position through the global
LGM (26e19 ka; Clark et al., 2009).
At the last glacial termination, the Mangatepopo glacier persisted
until 18e20 ka, as indicated by the cosmogenic 3
He moraine
chronology. Additional constraint from moraine tephrostratigraphy
in the Mangatepopo valley, suggests significant climatic amelioration
and glacier retreat occurred prior to deposition of the Waiohau
Tephra (c.14.0 ± 0.2 ka; Lowe et al., 2013). Several catchments in the
Southern Alps exhibit evidence for moraine formation at c.
20e22 ka and c. 18 ka (Schaefer et al., 2006; Putnam et al., 2013b;
Kelley et al., 2014). It is possible that a similar moraine sequence is
preserved in the Mangatepopo valley, but remains undated. For
example, we note that multiple, sharp-crested moraines exist on
Fig. 7. (a) Area-altitude curve for the reconstructed LGM Mangatepopo glacier. Solid
line depicts the associated ELA derived using an AAR of 0.67. Grey shading indicates
range using the AAR values of 0.5e0.8 (Meier and Post, 1962). (b) Cumulative probability
distribution function of the palaeotemperature estimate in the Mangatepopo
valley derived using palaeo- (1400e1550 m asl; AAR and MELM) and contemporaryELAs
(2483 ± 55 m asl; Keys, 1988) and a range of possible atmospheric temperature
lapse rates, equally weighted between 4 and 7

C kmÀ1
. Shaded grey zone indicates 1s
uncertainty interval.
S.R. Eaves et al. / Quaternary Science Reviews 132 (2016) 40e5652
the southern side of Mangatepopo valley (Fig. 2). However, thick
overlying soil sequences at these lower altitudes preclude exposure
dating of these moraines.
The palaeoglacier reconstruction for Mangatepopo valley indicates
local atmospheric temperature was reduced by 5.6 ± 1.1

C
relative to present (Fig. 7b) during the LGM. This estimate assumes
that precipitation during the LGM was similar to present. Currently
there is a paucity of quantitative LGM precipitation estimates in
New Zealand, however proxy-, glacier model-, and climate modelbased
reconstructions are generally consistent in suggesting that
annual precipitation was similar or slightly reduced, relative to
present (Drost et al., 2007; Rojas et al., 2009; Whittaker et al., 2011;
Golledge et al., 2012; Lorrey et al., 2012; Stephens et al., 2012).
Reduced annual precipitation in central North Island during the
LGM would increase the magnitude of atmospheric cooling
required to explain the reconstructed ELA. However, empirical and
glacier model-based evidence suggests that past and present-day
glacier mass balance in New Zealand is relatively insensitive to
precipitation change, with precipitation increases of c. 30e80%
required to balance 1

C of warming (Oerlemans, 1997; Anderson
and Mackintosh, 2006, 2012; Anderson et al., 2010). Thus, the error
in the palaeotemperature estimate arising from past precipitation
change is relatively insignificant (<1

C) in comparison to the
uncertainty in the ELA reconstruction and temperature lapse rate.
Our LGM temperature estimate for central North Island exhibits
good agreement with catchment- and regional-scale glacier model
simulations of LGM glaciers in the Southern Alps, which indicate
temperature depression of c. 6e7

C relative to present (Golledge
et al., 2012; McKinnon et al., 2012; Putnam et al., 2013b; Rowan
et al., 2013). The estimate also agrees with quantitative LGM temperature
reconstructions from dissolved noble gases in groundwater
(3.7e6.2

C; Seltzer et al., 2015) and fossil pollen assemblages
(6.0 ± 1.9

C e Newnham et al., 2013). Offshore, estimates of local
sea surface palaeotemperatures vary spatially, but also tend to
converge on a 4e7

C lowering between c. 30e18 ka (Pahnke and
Sachs, 2006; Barrows et al., 2007; Bostock et al., 2013).
5.2. Pre-LGM glaciation
Cosmogenic 3
He surface exposure ages from moraine M3 in
Mangatepopo valley indicate that the former Mangatepopo glacier
attained its maximum extent of the last glacial cycle prior to the
LGM. Surface erosion and tectonic uplift of the samples from the
M3 moraine, which have been exposed since at least c. 57 ka, means
that the exposure ages are likely to represent minimum ages for
moraine formation. Thus, the true moraine age is probably older,
perhaps by several millenia. Although undated, the moraine sequences
in the nearby Makahikatoa Stream catchment (Fig. 2)
exhibit similar morphology and morphostratigraphic relationships
to those in Mangatepopo valley, which further supports our interpretation
of a widespread glacier advance on Tongariro massif,
which occurred prior to the LGM. Below, we consider this interpretation
in the context of other geological records of glacial fluctuations
during this time period.
Elsewhere in New Zealand, evidence for pre-LGM glacial
expansion during the last glacial cycle remains relatively scarce.
Where present, the timing is often poorly constrained. However, a
recent high-precision cosmogenic 10
Be exposure age dataset from
the Balmoral moraines in central Southern Alps shows that an
advance of the former Pukaki and Tekapo glaciers culminated at
65 ± 3 ka (n ¼ 39, plus three outliers; Schaefer et al., 2015). The
Balmoral moraines associated with this advance (Barrell, 2014) are
present outboard of the well-dated, local LGM limits (Schaefer et al.,
2006; Kelley et al., 2014; Schaefer et al., 2015), thus indicating that
the glaciers attained their maximum extent of the last glacial cycle
prior to the LGM. Sutherland et al. (2007) present cosmogenic 10
Be
surface exposure ages from a suite of moraines preserved on
Cascade plateau (44

S), on the west coast of South Island. The
moraine belt immediately outboard of the LGM limits at this
location yielded boulder exposure ages of c. 60e68 ka (n ¼ 2, plus
one outlier) when recalculated according to Putnam et al. (2010b).
These revised ages provide further evidence for a glacial advance
during MIS 4, which was of similar magnitude to the LGM.
Other glacial records from South Island also indicate glacier
expansion around this time, however the chronologies are lessprecisely
constrained, largely relying on bracketing luminescence
ages from ice-marginal deposits. For example, McCarthy et al.
(2008) present evidence from the Tasman Mountains in northern
South Island (41

S) that suggests two periods of glaciation, of
similar magnitude, occurred during MIS 4 and the LGM. The MIS 4
glaciation at this location is constrained with optically-stimulated
luminescence ages of glacio-lacustrine deposits, which are present
outboard of the LGM limits, and date to 64 ± 10 ka. On the
central west coast of South Island, luminescence ages of sand/silt
beds interbedded with glacial outwash gravels suggest glacier
expansion at c. 85 ka and c. 64 ka (Preusser et al., 2005). However,
moraines correlated with these outwash deposits have recently
been shown to date to c. 25 ka (Barrows et al., 2013). No moraines of
MIS 4 age were recognised by Barrows et al. (2013), therefore they
suggest that any such deposits were overrun during the LGM when
glaciers in this catchment attained their maximum extent of the
last glacial cycle.
There is little evidence for any significant glacier advances in
New Zealand between 60 and 45 ka (Williams et al., 2015), during
which period fall most of the exposure ages from M3 moraine, if
taken at face value. Furthermore, continuous climate proxy data
indicate that this time period was characterised by relatively mild,
interstadial conditions (Shulmeister et al., 2001; Shane and
Sandiford, 2003; Whittaker et al., 2011; Williams et al., 2015),
which were probably unfavourable for significant glacier advance.
Offshore, Barrows et al. (2007) identify intervals of high clastic
sediment input between 70 and 60 ka in multiple cores, which are
intepreted to represent higher terrestrial erosion rates due to
expansion of nearby mountain glaciers. This is followed by a sharp
reduction after 60 ka, which is consistent with the terrestrial evidence
for climatic amelioration and glacial retreat. Thus, the majority
of evidence from glacial records and other climate proxies
support our interpretation that the M3 moraine is likely to represent
glacier advance in central North Island during late MIS 4 (c.
65e60 ka).
6. Conclusions
1. Tongariro massif was last glaciated between 30 and 18 ka when
a central ice field fed valley glaciers that extended down to c.
1200 m asl. The onset of glacial retreat occurred c. 18e20 ka,
which is in agreement with the only other moraine ages in
North Island (Brook et al., 2008). The timing of glaciation in
central North Island is also consistent with a growing body of
evidence that shows mountain glaciers in Southern Alps
attained their maximum Last Glacial Maximum extents by
32e28 ka and persisted until the onset of the last glacial
termination (Schaefer et al., 2006; Putnam et al., 2013b; Kelley
et al., 2014; Rother et al., 2014; Schaefer et al., 2015).
2. During the Last Glacial Maximum, the local equilibrium line
altitude in central North Island was c.1400e1550 m asl, which is
930e1080 m lower than present. This equates to a best-estimate
temperature depression of 5.6 ± 1.1

C, when uncertainties in the
ELA reconstruction and temperature lapse rate are considered.
This is good agreement with palaeotemperature estimates from
S.R. Eaves et al. / Quaternary Science Reviews 132 (2016) 40e56 53
glacier modelling (Golledge et al., 2012; McKinnon et al., 2012;
Putnam et al., 2013b) and other climate proxy reconstructions
(Newnham et al., 2013; Seltzer et al., 2015).
3. Reinvestigation of the moraine coverbed stratigraphy, using
field observations and major element analysis, indicates that the
rhyolitic tephra close to the moraine surface is the Waiohau
Tephra. This presence of this well-dated tephra within a palaeosol
overlying the innermost moraine indicates that climatic
amelioration following the LGM was well advanced in central
North Island by c. 14 cal. ka BP.
4. Cosmogenic 3
He surface exposure ages from boulders on the
outermost lateral moraine identified in the Mangatepopo valley
indicate that glaciers on Tongariro massif attained their greatest
extent of the last glacial cycle prior to the global Last Glacial
Maximum. Geological processes such as boulder surface erosion
and tectonic uplift mean that these ages provide minimum
limiting constraint of the moraine age and the true age is likely
older by several millennia. This evidence, together with comparison
to glacier records from Southern Alps (e.g. Schaefer
et al., 2015) and other continuous climate proxy data (e.g.
Barrows et al., 2007; Williams et al., 2015), indicates that this
glacial advance most likely occurred late in Marine Isotope Stage
4 (c. 65e60 ka).
Acknowledgements
S.R.E. was supported by the Victoria University Doctoral Scholarship,
a VUW Faculty Strategic Research Grant and the Antarctic
Research Centre Endowed Development Fund. J.M.S. and G.W.
acknowledge support by the Lamont Climate Center and the Comer
Science and Educational Foundation as well as by the NSF-EAR
award No. 0823521. Jenni Hopkins and Richard Jones assisted
with fieldwork. Roseanne Schwartz, Linda Baker and Sascha Serno
provided invaluable assistance with sample preparation and mass
spectrometry. We also thank Fred Luiszer (University of Colorado)
for major and trace element analyses. Permission for sample
collection was granted by Dept. of Conservation contract DM-
593774. This is LDEO contribution number 7947.
Appendix A. Supplementary data
Supplementary data related to this article can be found at http://
dx.doi.org/10.1016/j.quascirev.2015.11.004.
References
Almond, P., Moar, N., Lian, O., 2001. Reinterpretation of the glacial chronology of
South Westland, New Zealand. N. Z. J. Geol. Geophys. 44, 1e15.
Anderson, B., Mackintosh, A., 2006. Temperature change is the major driver of lateglacial
and Holocene glacier fluctuations in New Zealand. Geology 34, 121e124.
Anderson, B., Mackintosh, A., 2012. Controls on mass balance sensitivity of maritime
glaciers in the Southern Alps, New Zealand: the role of debris cover. J. Geophys.
Res. Earth Surf. 117, F01003.
Anderson, B., Mackintosh, A., Stumm, D., George, L., Kerr, T., Winter Billington, A.,
Fitzsimons, S., 2010. Climate sensitivity of a high-precipitation glacier in New
Zealand. J. Glaciol. 56, 114e128.
Andrews, J., 1975. Glacial Systems. An Approach to Glaciers and their Environments.
Duxbury Press, North Scituate, USA.
Andrews, J., 1985. The isotopic composition of radiogenic helium and its use to
study groundwater movement in confined aquifers. Chem. Geol. 49, 339e351.
Balco, G., Stone, J.O., Lifton, N.A., Dunai, T.J., 2008. A complete and easily accessible
means of calculating surface exposure ages or erosion rates from 10
Be and 26
Al
measurements. Quat. Geochronol. 4, 93e107.
Barrell, D.J.A., 2014. The Balmoral moraines near Lake Pukaki, Southern Alps: a new
reference area for the early Otira Glaciation in New Zealand. N. Z. J. Geol.
Geophys. 57, 442e452.
Barrell, D.J.A., Almond, P.C., Vandergoes, M.J., Lowe, D.J., Newnham, R.M., 2013.
A composite pollen-based stratotype for inter-regional evaluation of climatic
events in New Zealand over the past 30,000 years (NZ-INTIMATE project). Quat.
Sci. Rev. 74, 4e20.
Barrows, T.T., Almond, P., Rose, R., Keith Fifield, L., Mills, S.C., Tims, S.G., 2013. Late
Pleistocene glacial stratigraphy of the Kumara-Moana region, West Coast of
South Island, New Zealand. Quat. Sci. Rev. 74, 139e159.
Barrows, T.T., Juggins, S., De Deckker, P., Calvo, E., Pelejero, C., 2007. Long-term sea
surface temperature and climate change in the Australian-New Zealand region.
Paleoceanography 22, PA2215.
Benn, D.I., Owen, L.A., Osmaston, H.A., Seltzer, G.O., Porter, S.C., Mark, B., 2005.
Reconstruction of equilibrium-line altitudes for tropical and sub-tropical glaciers.
Quat. Int. 138e139, 8e21.
Bilderback, E.L., 2012. Hillslope Response to Climate-modulated River Incision and
the Role of Deep-seated Landslides in Post-glacial Sediment Flux: Waipaoa
Sedimentary System, New Zealand. Ph.D. Thesis. University of Canterbury, New
Zealand.
Blard, P.H., Farley, K.A., 2008. The influence of radiogenic 4He on cosmogenic 3
He
determinations in volcanic olivine and pyroxene. Earth Planet. Sci. Lett. 276,
20e29.
Borchers, B., Marrero, S., Balco, G., Caffee, M., Goehring, B., Lifton, N., Nishiizumi, K.,
Phillips, F., Schaefer, J., Stone, J., 2015. Geological calibration of spallation production
rates in the cronus-earth project. Quat. Geochronol. http://dx.doi.org/
10.1016/j.quageo.2015.09.005.
Bostock, H., Barrows, T., Carter, L., Chase, Z., Cortese, G., Dunbar, G., Ellwood, M.,
Hayward, B., Howard, W., Neil, H., Noble, T., Mackintosh, A., Moss, P., Moy, A.,
White, D., Williams, M., Armand, L., 2013. A review of the AustralianNew Zealand
sector of the Southern Ocean over the last 30 ka (Aus-INTIMATE project).
Quat. Sci. Rev. 74, 35e57.
Briner, J.P., 2009. Moraine pebbles and boulders yield indistinguishable 10
Be ages: a
case study from Colorado, USA. Quat. Geochronol. 4, 299e305.
Bromley, G.R., Winckler, G., Schaefer, J.M., Kaplan, M.R., Licht, K.J., Hall, B.L., 2014.
Pyroxene separation by HF leaching and its impact on helium surface-exposure
dating. Quat. Geochronol. 23, 1e8.
Brook, M.S., 2009. Lateral moraine age in Park Valley, Tararua Range, New Zealand.
J. R. Soc. N. Z. 39, 63e69.
Brook, M.S., Crow, T.V.H., 2008. A debris ridge in Park Valley, Tararua Range, New
Zealand as evidence for Pleistocene glaciation. N. Z. J. Geol. Geophys. 51, 23e28.
Brook, M.S., Dean, J.F., Keys, H.J.R., 2011. Response of a mid-latitude cirque glacier to
climate over the last two decades: Mangaehuehu Glacier, Mt Ruapehu. Earth
Surf. Process. Landf. 36, 1973e1980.
Brook, M.S., Shulmeister, J., Crow, T.V.H., Zondervan, A., 2008. First cosmogenic 10
Be
constraints on LGM glaciation on New Zealand's North Islands: Park Valley,
Tararua Range. J. Quat. Sci. 23, 707e712.
Chinn, T., Fitzharris, B.B., Willsman, A., Salinger, M.J., 2012. Annual ice volume
changes 1976e2008 for the New Zealand Southern Alps. Glob. Planet. Change
92e93, 105e118.
Clark, P.U., Dyke, A.S., Shakun, J.D., Carlson, A.E., Clark, J., Wohlfarth, B.,
Mitrovica, J.X., Hostetler, S.W., McCabe, A.M., 2009. The Last Glacial Maximum.
Science 325, 710e714.
Columbus, J., Sirguey, P., Tenzer, R., 2011. A free, fully accessible 15-m DEM for New
Zealand. Surv. Q. 66, 16e19.
Cronin, S.J., Neall, V.E., 1997. A late quaternary stratigraphic framework for the
northeastern Ruapehu and eastern Tongariro ring plains, New Zealand. N. Z. J.
Geol. Geophys. 40, 185e197.
Denton, G.H., Anderson, R.F., Toggweiler, J.R., Edwards, R.L., Schaefer, J.M.,
Putnam, A.E., 2010. The last glacial termination. Science 328, 1652e1656.
Desilets, D., Zreda, M., Prabu, T., 2006. Extended scaling factors for in situ cosmogenic
nuclides: new measurements at low latitude. Earth Planet. Sci. Lett. 246,
265e276.
Donoghue, S.L., Neall, V.E., Palmer, A.S., 1995. Stratigraphy and chronology of late
Quaternary andesitic tephra deposits, Tongariro volcanic Centre, New Zealand.
J. R. Soc. N. Z. 25, 115e206.
Doughty, A.M., Anderson, B.M., Mackintosh, A.N., Kaplan, M.R., Vandergoes, M.J.,
Barrell, D.J.A., Denton, G.H., Schaefer, J.M., Chinn, T.J.H., Putnam, A.E., 2013.
Evaluation of Lateglacial temperatures in the Southern Alps of New Zealand
based on glacier modelling at Irishman Stream, Ben Ohau Range. Quat. Sci. Rev.
74, 160e169.
Drost, F., Renwick, J., Bhaskaran, B., Oliver, H., McGregor, J., 2007. A simulation of
New Zealand's climate during the Last Glacial Maximum. Quat. Sci. Rev. 26,
2505e2525.
Dunai, T.J., 2001. Influence of secular variation of the geomagnetic field on production
rates of in situ produced cosmogenic nuclides. Earth Planet. Sci. Lett.
193, 197e212.
Dunne, J., Elmore, D., Muzikar, P., 1999. Scaling factors for the rates of production of
cosmogenic nuclides for geometric shielding and attenuation at depth on
sloped surfaces. Geomorphology 27, 3e11.
Eaves, S.R., Winckler, G., Schaefer, J.M., Vandergoes, M.J., Alloway, B.V.,
Mackintosh, A.N., Townsend, D.B., Ryan, M.T., Li, X., 2015. A test of the cosmogenic
3
He production rate in the south west Pacific (39

S). J. Quat. Sci. 30,
87e97.
Froggatt, P.C., Lowe, D.J., 1990. A review of late Quaternary silicic and some other
tephra formations from New Zealand: their stratigraphy, nomenclature, distribution,
volume, and age. N. Z. J. Geol. Geophys. 33, 89e109.
Furbish, D.J., Andrews, J.T., 1984. The use of hypsometry to indicate long-term
stability and response of valley glaciers to changes in mass transfer. J. Glaciol.
30, 199e211.
Goehring, B.M., Kurz, M.D., Balco, G., Schaefer, J.M., Licciardi, J., Lifton, N., 2010.
A reevaluation of in situ cosmogenic 3
He production rates. Quat. Geochronol. 5,
410e418.
S.R. Eaves et al. / Quaternary Science Reviews 132 (2016) 40e5654
Golledge, N.R., Mackintosh, A.N., Anderson, B.M., Buckley, K.M., Doughty, A.M.,
Barrell, D.J.A., Denton, G.H., Vandergoes, M.J., Andersen, B.G., Schaefer, J.M.,
2012. Last Glacial Maximum climate in New Zealand inferred from a modelled
Southern Alps icefield. Quat. Sci. Rev. 46, 30e45.
Hobden, B.J., Houghton, B.F., Lanphere, M.A., Nairn, I.A., 1996. Growth of the Tongariro
volcanic complex: new evidence from K-Ar age determinations. N. Z. J.
Geol. Geophys. 39, 151e154.
Hulston, J.R., Lupton, J.E., 1996. Helium isotope studies of geothermal fields in the
Taupo Volcanic Zone, New Zealand. J. Volcanol. Geotherm. Res. 74, 297e321.
Ivy-Ochs, S., Kerschner, H., Schlchter, C., 2007. Cosmogenic nuclides and the dating
of Lateglacial and Early Holocene glacier variations: the Alpine perspective.
Quat. Int. 164165, 53e63.
Kaplan, M.R., Schaefer, J.M., Denton, G.H., Barrell, D.J.A., Chinn, T.J.H., Putnam, A.E.,
Andersen, B.G., Finkel, R.C., Schwartz, R., Doughty, A.M., 2010. Glacier retreat in
New Zealand during the Younger Dryas Stadial. Nature 467, 194e197.
Kaplan, M.R., Schaefer, J.M., Denton, G.H., Doughty, A.M., Barrell, D.J.A., Chinn, T.J.H.,
Putnam, A.E., Andersen, B.G., Mackintosh, A., Finkel, R.C., et al., 2013. The
anatomy of long-term warming since 15 ka in New Zealand based on net glacier
snowline rise. Geology 41, 887e890.
Kelley, S.E., Kaplan, M.R., Schaefer, J.M., Andersen, B.G., Barrell, D.J., Putnam, A.E.,
Denton, G.H., Schwartz, R., Finkel, R.C., Doughty, A.M., 2014. High-precision 10
Be chronology of moraines in the Southern Alps indicates synchronous cooling
in Antarctica and New Zealand 42,000 years ago. Earth Planet. Sci. Lett. 405,
194e206.
Keys, H., 1988. Survey of the Glaciers on Mt. Ruapehu, Tongariro National Park e a
Baseline for Detecting Effects of Climate Change. Department of Conservation,
Wellington, NZ, p. 1988. Technical Report Science and Research Internal Report
No.24.
Kirkbride, M.P., Dugmore, A.J., 2001. Timing and significance of mid-Holocene
glacier advances in northern and central Iceland. J. Quat. Sci. 16, 145e153.
Kuhn, M., 1995. The mass balance of very small glaciers. Z. für Gletscherkd. Glazialgeol.
31, 171e179.
Lal, D., 1991. Cosmic ray labeling of erosion surfaces: in situ nuclide production rates
and erosion models. Earth Planet. Sci. Lett. 104, 424e439.
Licciardi, J.M., Clark, P.U., Brook, E.J., Elmore, D., Sharma, P., 2004. Variable responses
of western U.S. glaciers during the last deglaciation. Geology 32, 81e84.
Lifton, N., Sato, T., Dunai, T.J., 2014. Scaling in situ cosmogenic nuclide production
rates using analytical approximations to atmospheric cosmic-ray fluxes. Earth
Planet. Sci. Lett. 386, 149e160.
Lifton, N., Smart, D.F., Shea, M.A., 2008. Scaling time-integrated in situ cosmogenic
nuclide production rates using a continuous geomagnetic model. Earth Planet.
Sci. Lett. 268, 190e201.
Lorrey, A.M., Vandergoes, M., Almond, P., Renwick, J., Stephens, T., Bostock, H.,
Mackintosh, A., Newnham, R.M., Williams, P.W., Ackerley, D., Neil, H.,
Fowler, A.M., 2012. Palaeocirculation across New Zealand during the last glacial
maximum at 21 ka. Quat. Sci. Rev. 36, 189e213.
Lowe, D.J., Blaauw, M., Hogg, A.G., Newnham, R.M., 2013. Ages of 24 widespread
tephras erupted since 30,000 years ago in New Zealand, with re-evaluation of
the timing and palaeoclimatic implications of the Lateglacial cool episode
recorded at Kaipo bog. Quat. Sci. Rev. 74, 170e194.
Lowe, D.J., Shane, P.A.R., Alloway, B.V., Newnham, R.M., 2008. Fingerprints and age
models for widespread New Zealand tephra marker beds erupted since 30,000
years ago: a framework for NZ-INTIMATE. Quat. Sci. Rev. 27, 95e126.
Marden, M., Betts, H., Palmer, A., Taylor, R., Bilderback, E., Litchfield, N., 2014. PostLast
Glacial Maximum fluvial incision and sediment generation in the unglaciated
Waipaoa catchment, North Island, New Zealand. Geomorphology 214,
283e306.
Marrero, S.M., Phillips, F.M., Borchers, B., Lifton, N., Aumer, R., Balco, G., 2015.
Cosmogenic nuclide systematics and the cronuscalc program. Quat. Geochronol.
http://dx.doi.org/10.1016/j.quageo.2015.09.005.
Mathews, W.H., 1967. A contribution to the geology of the Mount Tongariro massif,
North Island, New Zealand. N. Z. J. Geol. Geophys. 10, 1027e1038.
McArthur, J.L., Shepherd, M.J., 1990. Late Quaternary glaciation of Mt Ruapehu,
North Island, New Zealand, New Zealand. J. R. Soc. N. Z. 20, 287e296.
McCarthy, A., Mackintosh, A., Rieser, U., Fink, D., 2008. Mountain glacier chronology
from Boulder Lake, New Zealand, indicates MIS 4 and MIS 2 ice advances of
similar extent. Arct. Antarct. Alp. Res. 40, 695e708.
McKinnon, K.A., Mackintosh, A.N., Anderson, B.M., Barrell, D.J.A., 2012. The influence
of sub-glacial bed evolution on ice extent: a model-based evaluation of the
Last Glacial Maximum Pukaki glacier, New Zealand. Quat. Sci. Rev. 57, 46e57.
Meier, M.F., Post, A.S., 1962. Recent Variations in Mass Net Budgets of Glaciers in
Western North America. International Association of Scientific Hydrology
Publication, pp. 63e77, 58.
Newnham, R.M., McGlone, M., Moar, N., Wilmshurst, J., Vandergoes, M., 2013. The
vegetation cover of New Zealand at the Last Glacial Maximum. Quat. Sci. Rev.
74, 202e214.
Nishiizumi, K., Winterer, E., Kohl, C., Klein, J., Middleton, R., Lal, D., Arnold, J., 1989.
Cosmic ray production rates of 10
Be and 26
Al in quartz from glacially polished
rocks. J. Geophys. Res. Solid Earth 1978e2012 (94), 17907e17915.
NIWA, 2014. CliFlo: NIWA's National Climate Database on the Web. Website. http://
cliflo.niwa.co.nz/. Retrieved 2014.
Norton, D.A., 1985. A multivariate technique for estimating New Zealand temperature
normals. Weather Clim. 5, 64e74.
Oerlemans, J., 1997. Climate sensitivity of Franz Josef Glacier, New Zealand, as
revealed by numerical modeling. Arct. Alp. Res. 29, 233e239.
Oerlemans, J., Fortuin, J.P.F., 1992. Sensitivity of glaciers and small ice caps to
greenhouse warming. Science 258, 115e117.
Osmaston, H., 2005. Estimates of glacier equilibrium line altitudes by the Area
Altitude, the Area Altitude Balance Ratio and the Area Altitude Balance Index
methods and their validation. Quat. Int. 138e139, 22e31.
Pahnke, K., Sachs, J.P., 2006. Sea surface temperatures of southern midlatitudes
0e160 kyr BP. Paleoceanography 21, PA2003.
Paterson, W.S.B., 1994. The Physics of Glaciers. Elsevier, Oxford, New York and
Tokyo.
Patterson, D., Honda, M., McDougall, I., 1994. Noble gases in mafic phenocrysts and
xenoliths from New Zealand. Geochimica Cosmochimica Acta 58, 4411e4427.
Portenga, E.W., Bierman, P.R., 2011. Understanding Earth's eroding surface with 10
Be. GSA Today 21, 4e10.
Porter, S.C., 1975. Equilibrium-line altitudes of late Quaternary glaciers in the
Southern Alps, New Zealand. Quat. Res. 5, 27e47.
Preusser, F., Andersen, B.G., Denton, G.H., Schlüchter, C., 2005. Luminescence
chronology of Late Pleistocene glacial deposits in North Westland, New Zealand.
Quat. Sci. Rev. 24, 2207e2227.
Pulford, A.K., 2002. Crustal Structure and Lithospheric Doming: Aspects of Deformation
along an Obliquely Convergent Plate Margin, New Zealand. Ph.D. Thesis.
Victoria University of Wellington.
Purdie, H., Anderson, B., Chinn, T., Owens, I., Mackintosh, A., Lawson, W., 2014. Franz
Josef and Fox Glaciers, New Zealand: historic length records. Glob. Planet.
Change 121, 41e52.
Putnam, A.E., Denton, G.H., Schaefer, J.M., Barrell, D.J.A., Andersen, B.G., Finkel, R.C.,
Schwartz, R., Doughty, A.M., Kaplan, M.R., Schlüchter, C., 2010a. Glacier advance
in southern middle-latitudes during the Antarctic Cold Reversal. Nat. Geosci. 3,
700e704.
Putnam, A.E., Schaefer, J.M., Barrell, D.J.A., Vandergoes, M., Denton, G.H.,
Kaplan, M.R., Finkel, R.C., Schwartz, R., Goehring, B.M., Kelley, S.E., 2010b. In situ
cosmogenic 10
Be production-rate calibration from the Southern Alps, New
Zealand. Quat. Geochronol. 5, 392e409.
Putnam, A.E., Schaefer, J.M., Denton, G.H., Barrell, D.J.A., Andersen, B.G.,
Koffman, T.N.B., Rowan, A.V., Finkel, R.C., Rood, D.H., Schwartz, R.,
Vandergoes, M.J., Plummer, M.A., Brocklehurst, S.H., Kelley, S.E., Ladig, K.L.,
2013a. Warming and glacier recession in the Rakaia valley, Southern Alps of
New Zealand, during Heinrich Stadial 1. Earth Planet. Sci. Lett. 382, 98e110.
Putnam, A.E., Schaefer, J.M., Denton, G.H., Barrell, D.J.A., Birkel, S.D., Andersen, B.G.,
Kaplan, M.R., Finkel, R.C., Schwartz, R., Doughty, A.M., 2013b. The Last Glacial
Maximum at 44

S documented by a 10
Be moraine chronology at Lake Ohau,
Southern Alps of New Zealand. Quat. Sci. Rev. 62, 114e141.
Putnam, A.E., Schaefer, J.M., Denton, G.H., Barrell, D.J.A., Finkel, R.C., Andersen, B.G.,
Schwartz, R., Chinn, T.J., Doughty, A.M., 2012. Regional climate control of glaciers
in New Zealand and Europe during the pre-industrial Holocene. Nat.
Geosci. 5, 627e630.
Rea, B.R., Whalley, W.B., Dixon, T.S., Gordon, J.E., 1999. Plateau icefields as
contributing areas to valley glaciers and the potential impact on reconstructed
ELAs: a case study from the Lyngen Alps, North Norway. Ann. Glaciol. 28,
97e102.
Roethlisberger, F., Schneebeli, W., 1979. Genesis of lateral moraine complexes,
demonstrated by fossil soils and trunks; indicators of postglacial climatic
fluctuations. In: Schlüchter, C. (Ed.), Moraines and Varves: Origin, Genesis,
Classification. Balkema, Rotterdam, The Netherlands, pp. 387e420.
Rojas, M., Moreno, P., Kageyama, M., Crucifix, M., Hewitt, C., Abe Ouchi, A.,
Ohgaito, R., Brady, E.C., Hope, P., 2009. The Southern Westerlies during the last
glacial maximum in PMIP2 simulations. Clim. Dyn. 32, 525e548.
Rother, H., Fink, D., Shulmeister, J., Mifsud, C., Evans, M., Pugh, J., 2014. The early rise
and late demise of New Zealand's last glacial maximum. Proc. Natl. Acad. Sci.
111, 11630e11635.
Rowan, A.V., Plummer, M.A., Brocklehurst, S.H., Jones, M.A., Schultz, D.M., 2013.
Drainage capture and discharge variations driven by glaciation in the Southern
Alps, New Zealand. Geology 41, 199e202.
Schaefer, J.M., Denton, G.H., Barrell, D.J.A., Ivy Ochs, S., Kubik, P.W., Andersen, B.G.,
Phillips, F.M., Lowell, T.V., Schlüchter, C., 2006. Near-synchronous interhemispheric
termination of the Last Glacial Maximum in mid-latitudes. Science 312,
1510e1513.
Schaefer, J.M., Denton, G.H., Kaplan, M., Putnam, A., Finkel, R.C., Barrell, D.J.A.,
Andersen, B.G., Schwartz, R., Mackintosh, A., Chinn, T., Schlüchter, C., 2009.
High-frequency Holocene glacier fluctuations in New Zealand differ from the
northern signature. Science 324, 622e625.
Schaefer, J.M., Putnam, A.E., Denton, G.H., Kaplan, M.R., Birkel, S., Doughty, A.M.,
Kelley, S., Barrell, D.J.A., Finkel, R.C., Winckler, G., Anderson, R.F., Ninneman, U.S.,
Barker, S., Schwartz, R., Schlüchter, C., 2015. The Southern Glacial Maximum
65000 years ago and its Unfinished Termination. Quat. Sci. Rev. 114, 52e60.
Seltzer, A., Stute, M., Morgenstern, U., Stewart, M., Schaefer, J., 2015. Mean annual
temperature in new zealand during the last glacial maximum derived from
dissolved noble gases in groundwater. Earth Planet. Sci. Lett. 431, 206e216.
Shane, P., 2000. Tephrochronology: a New Zealand case study. Earth Sci. Rev. 49,
223e259.
Shane, P., Nairn, I.A., Martin, S.B., Smith, V.C., 2008. Compositional heterogeneity in
tephra deposits resulting from the eruption of multiple magma bodies: implications
for tephrochronology. Quat. Int. 178, 44e53.
Shane, P., Sandiford, A., 2003. Paleovegetation of marine isotope stages 4 and 3 in
Northern New Zealand and the age of the widespread Rotoehu tephra. Quat.
Res. 59, 420e429.
S.R. Eaves et al. / Quaternary Science Reviews 132 (2016) 40e56 55
Shulmeister, J., Shane, P., Lian, O.B., Okuda, M., Carter, J.A., Harper, M., Dickinson, W.,
Augustinus, P., Heijnis, H., 2001. A long late-Quaternary record from Lake
Poukawa, Hawkes Bay, New Zealand. Palaeogeogr. Palaeoclimatol. Palaeoecol.
176, 81e107.
Sikes, E.L., Howard, W.R., Samson, C.R., Mahan, T.S., Robertson, L.G., Volkman, J.K.,
2009. Southern ocean seasonal temperature and subtropical front movement
on the South Tasman Rise in the Late Quaternary. Paleoceanography 24, PA2201.
Stephens, T., Atkin, D., Cochran, U., Augustinus, P., Reid, M., Lorrey, A., Shane, P.,
Street Perrott, A., 2012. A diatom-inferred record of reduced effective precipitation
during the Last Glacial Coldest Phase (28.8e18.0 cal kyr BP) and
increasing Holocene seasonality at Lake Pupuke, Auckland, New Zealand.
J. Paleolimnol. 48, 801e817.
Stone, J.O., 2000. Air pressure and cosmogenic isotope production. J. Geophys. Res. B
Solid Earth 105, 23753e23759.
Sturman, A.P., Tapper, N.J., 1996. The Weather and Climate of Australia and New
Zealand. Oxford University Press, Melbourne, Australia.
Suganuma, Y., Miura, H., Okuno, J., 2012. A new sampling technique for surface
exposure dating using a portable electric rock cutter. Antarct. Rec. 56, 85e90.
Suggate, R.P., 1990. Late Pliocene and quaternary glaciations of New Zealand. Quat.
Sci. Rev. 9, 175e197.
Suggate, R.P., Almond, P.C., 2005. The Last Glacial Maximum (LGM) in western South
Island, New Zealand: Implications for the global LGM and MIS 2. Quat. Sci. Rev.
24, 1923e1940.
Sutherland, R., Kim, K., Zondervan, A., McSaveney, M., 2007. Orbital forcing of midlatitude
Southern Hemisphere glaciation since 100 ka inferred from cosmogenic
nuclide ages of moraine boulders from the Cascade Plateau, southwest
New Zealand. Bull. Geol. Soc. Am. 119, 443e451.
Sutton, P.J.H., Bowen, M., Roemmich, D., 2005. Decadal temperature changes in the
Tasman Sea. N. Z. J. Mar. Freshw. Res. 39, 1321e1329.
Topping, W.W., 1974. Some Aspects of Quaternary History of Tongariro Volcanic
Centre. Ph.D. Thesis. Victoria University of Wellington.
Topping, W.W., Kohn, B.P., 1973. Rhyolitic tephra marker beds in the Tongariro area,
North Island, New Zealand. N. Z. J. Geol. Geophys. 16, 375e395.
Torgersen, T., Lupton, J., Sheppard, D., Giggenbach, W., 1982. Helium isotope variations
in the thermal areas of New Zealand. J. Volcanol. Geotherm. Res. 12,
283e298.
Uddstrom, M.J., Oien, N.A., 1999. On the use of high-resolution satellite data to
describe the spatial and temporal variability of sea surface temperatures in the
New Zealand region. J. Geophys. Res. C Oceans 104, 20729e20751.
Vandergoes, M.J., Hogg, A.G., Lowe, D.J., Newnham, R.M., Denton, G.H., Southon, J.,
Barrell, D.J.A., Wilson, C.J.N., McGlone, M.S., Allan, A.S.R., Almond, P.C.,
Petchey, F., Dabell, K., Dieffenbacher Krall, A.C., Blaauw, M., 2013. A revised age
for the Kawakawa/Oruanui tephra, a key marker for the Last Glacial Maximum
in New Zealand. Quat. Sci. Rev. 74, 195e201.
Whittaker, T.E., Hendy, C.H., Hellstrom, J.C., 2011. Abrupt millennial-scale changes in
intensity of Southern Hemisphere westerly winds during marine isotope stages
2e4. Geology 39, 455e458.
Williams, P.W., 1996. A 230 ka record of glacial and interglacial events from Aurora
Cave, Fiordland, New Zealand. N. Z. J. Geol. Geophys. 39, 225e241.
Williams, P.W., McGlone, M., Neil, H., Zhao, J.X., 2015. A review of New Zealand
palaeoclimate from the Last Interglacial to the global Last Glacial Maximum.
Quat. Sci. Rev. 110, 92e106.
Winckler, G., Anderson, R.F., Schlosser, P., 2005. Equatorial Pacific productivity and
dust flux during the mid-Pleistocene climate transition. Paleoceanography 20,
PA4025.
S.R. Eaves et al. / Quaternary Science Reviews 132 (2016) 40e5656