Investigation of late-Holocene moraines in the western Southern Alps,New Zealand,applying Schmidt-hammer exposure-age dating

Abstract

Chronological studies applying Schmidt-hammer exposure-age dating (SHD) were performed on six glacier forelands in the western part of the Southern Alps, New Zealand. Although lithological heterogeneity prevented a regional age-calibration curve to be established, local age-calibration curves for La Perouse Glacier and Strauchon Glacier could be derived. They show similar linear equations and trends/slopes and enabled a preliminary assessment of the representativeness of individual ¹⁰Be terrestrial cosmogenic nuclide dating (TCND)-ages obtained from the other forelands. No mid- and early-Holocene advance periods were detected. Clusters of moraine ages date around 2800, 1850–1450, and 1100–900 years ago, followed by the ‘Little Ice Age’ commencing c. 500 years ago. There is no good agreement with earlier radiocarbon-based studies and recently published TCND-chronologies. As it will be outlined, this partly could be the result of different approaches to the palaeoclimatic interpretation of the dated samples. The results obtained from this recent study do not support an elsewhere proposed general asynchronous glacier behaviour between the mid-latitudinal Northern and Southern hemispheres.

Keywords

glacier chronology Holocene New Zealand Schmidt-hammer exposure-age dating Southern Alps terrestrial cosmogenic nuclide dating

Introduction

The importance of mountain glaciers as key indicators for past and present climate change has frequently been highlighted (Intergovernmental Panel on Climate Change (IPCC), 2007; World Glacier Monitoring Service (WGMS), 2008; Zemp et al., 2009). Research on Holocene mountain glacier chronologies has accordingly intensified recently (cf. Davis et al., 2009; Grove, 2004; Orombelli, 2011; Solomina et al., 2008; Wanner et al., 2008, 2011). This progress is closely connected with major improvements in modern dating techniques like terrestrial cosmogenic nuclide dating (TCND), and the application of multi-proxy approaches adapted to specific local/regional conditions (Böhlert et al., 2011; Jomelli et al., 2011; Matthews and Owen, 2010; Putnam et al., 2010a, 2012; Schaefer et al., 2009; Shakesby et al., 2006, 2011). Despite recent progress, there is, however, still a need for better spatial differentiation (Winkler and Matthews, 2010a; Winkler et al., 2010). Detection of temporal and spatial patterns in Holocene glacier chronologies is essential to assess the underlying atmospheric circulation patterns and the influence of external versus internal factors on climate forcing. Another aspect recently elaborated by Kirkbride and Winkler (2012) is the lack of generally accepted guidelines or concepts for global and inter-hemispheric correlation of Holocene glacier chronologies. Such guidelines should provide an indispensable basis for any attempt to analyse climatic teleconnections and ocean–atmosphere couplings.

The Southern Alps of New Zealand have provided one of only a few suitable study sites for investigating Holocene glacier chronologies in the mid-latitudinal Southern Hemisphere (e.g. Burrows and Gellatly, 1982; Gellatly et al., 1988; Röthlisberger, 1986). Recent years have seen a significant increase in TCND ages published for the Southern Alps, mainly for the last glaciation (Barrell et al., 2011; Kaplan et al., 2010; Putnam et al., 2010a, 2013), but also for the Holocene (e.g. Putnam et al., 2012; Schaefer et al., 2009). These studies focus, as in the majority of previous work (e.g. Burrows, 1973, 1980, 1989; Gellatly, 1982, 1984, 1985a, 1985b; Röthlisberger, 1986; Winkler, 2000, 2004, 2005), primarily on selected glacier forelands east of the Main Divide in Mt Cook/Aoraki National Park. Bad accessibility and methodological problems account for comparatively few investigations on glacier foreland west of the Main Divide. The only exceptions are ‘Little Ice Age’ (LIA) and recent glacier variations at Franz Josef and Fox Glaciers (e.g. Anderson and Mackintosh, 2006; Anderson et al., 2008; Fitzharris et al., 1992a; Hooker and Fitzharris, 1999; McKinzey et al., 2004; Woo and Fitzharris, 1992). In addition, Wardle (1973, 1978), Röthlisberger (1986) and Burrows et al. (2002) provide some radiocarbon ages for Holocene moraines and glaciofluvial deposits in the West and, lately, Winkler (2009a) reported the first late-Holocene TCND ages from an individual glacier (Strauchon Glacier) in this region.

The major aims of this study can be summarised as following:

To present new data on the (late-) Holocene glacier chronology from six glaciers in the western part of the Southern Alps, combined with thorough re-assessing of previously reported moraine TCND ages from Strauchon Glacier (Winkler, 2009a) applying the new regional ¹⁰Be production rate (Putnam et al., 2010b);

To test the applicability of Schmidt-hammer exposure-age dating (SHD) in the Southern Alps and attempt to calculate a regional age-calibration curve;

To compare the results with previously published radiocarbon ages and TCND-based glacier chronologies (Putnam et al., 2012; Schaefer et al., 2009) in order to detect possible regional patterns of synchronous glacier variations during the late Holocene;

To attempt an interpretation of these results in a global palaeoclimatic context and outline methodological potential and restrictions.

Study area

The Southern Alps run roughly parallel to the west coast of New Zealand’s South Island. The Main Divide is relatively close to the west coast (minimum distance c. 20 km). A significant concentration of New Zealand’s glaciated area of, in total, c. 3000 glaciers occurs within the central and highest part of the Southern Alps around Mt Cook. The overall glaciated area in New Zealand was c. 1160 km² in the 1970s inventory (Chinn, 2001), but the ice volume has decreased by around 15% since then (Chinn et al., 2012). The Southern Alps are exposed to a maritime mid-latitudinal climate with a strong west-east precipitation gradient (Chinn et al., 2005). Values of 3000 mm annual precipitation for the narrow western coastal plains rise to at least 10,000 mm over the ranges and névés west of the Main Divide (Chinn and Whitehouse, 1980). Precipitation decreases sharply towards the eastern ranges, but the glaciated area just east of the Main Divide still can be characterised as maritime (Chinn et al., 2008; Winkler et al., 2010). A typical feature of many New Zealand glaciers is a considerable supraglacial debris cover (Anderson and Mackintosh, 2012; Chinn, 1996). This debris cover and the proglacial lakes developed at several valley glaciers during the 20th century (Chinn et al., 2005, 2008) have to be considered in any interpretation of glacier behaviour (Kirkbride and Winkler, 2012; Reznichenko et al., 2012; WGMS, 2008; Winkler et al., 2010).

Five of the six glaciers studied (Balfour, Horace Walker, La Perouse, Marchant and Strauchon) are located in Westland/Tai Poutini National Park (Figure 1). Only Whymper Glacier in the headwaters of the Whataroa Valley is outside its boundary. Although exhibiting different sizes and glaciological properties (Table 1), all glaciers are single-basin glaciers and quite representative for valley glaciers of small- and medium-size in the Southern Alps. Except for Horace Walker Glacier, all are characterised by a widespread supraglacial debris cover on the lower tongues. Most of those debris-covered lower tongues appear to be stagnant, and at several glaciers (e.g. Balfour Glacier and La Perouse Glacier), a proglacial lake is currently under development.

Figure 1.

Topographic sketch maps of the study area: (a) the location of the detailed map (see insert) and (b) the central part of the Southern Alps around Mt Cook and the six glaciers studied (in italics).

Table 1.

Glaciological key data of the six glaciers studied. The data (late 1970s inventory) were taken from Chinn (1996) and World Glacier Inventory (http://www.nsid.org/data/glacier_inventory/). Parameters will have changed during the past c. 35 years, but no updated inventory is available yet. In the case of discrepancy between the data sources, Chinn (1996) has been followed.

	Area (km²)	Length (km)	Altitude (m a.s.l.)	Aspect^a (accumulation − ablation)	Debris cover (%)
Balfour Glacier	7	9.8	3305–730	W − W	48
Horace Walker Glacier	5.99	6.5	2455–945	W − SW	0
La Perouse Glacier	9.5	11.6	3320–855	NW − W	30
Marchant Glacier	1.3	2.95	2255–1065	SW − W	52
Strauchon Glacier	4.17	5.9	2530–960	W – SW	41
Whymper Glacier	6.56	7.2	2775–790	NW – NE	58

Aspect given for accumulation and ablation area.

The main factor for the selection of these glaciers was the presence of potential sites for TCND sampling and Schmidt-hammer testing (Figure 2). Limited accessibility and time constraints meant that only selected parts of the forelands and moraines/moraine systems could be examined on the ground. Main criterion for the selection of sampling/test sites was the abundance of suitable boulders for Schmidt-hammer testing. Sites above the local tree line or with sparse scrub vegetation were preferred. Sections of lateral moraines potentially affected by postdepositional erosion were avoided. The fieldwork focussed on ‘alpine type’ lateral moraines sensu Winkler and Hagedorn (1999), mainly on multi-ridged lateral moraine systems built up in accretion mode sensu Röthlisberger and Schneebeli (1979; cf. Winkler, 2009b). This moraine type is regarded as most suitable for surface exposure dating and probably less exposed to the influence of large-scale mass movements (cf. Larsen et al., 2005; Reznichenko et al., 2011, 2012; Shulmeister et al., 2009; Vacco et al., 2010).

Figure 2.

Glacier forelands studied: (a) Balfour Glacier and foreland (1 March 2007). The up-valley section of its southern lateral moraine is framed. (b) Lateral moraine at Balfour Glacier looking down-glacier (5 April 2008). (c) Horace Walker Glacier and (d) erosion scarp on the proximal side at the innermost ridge of its prominent moraine system (2 March 2007). (e) Lower glacier tongue of La Perouse Glacier (1 March 2007). The northern lateral moraine section investigated is framed. (f) View down-glacier from this northern lateral moraine section towards lateral and latero-frontal ridges in the southern foreland (3 April 2008). (g) Marchant Glacier and its foreland (30 March 2008). The south-western foreland investigated is framed. (h) Looking east from the outermost moraine ridge at Marchant Glacier showing efficient paraglacial erosion that has already destroyed most moraines in the inner foreland. Note the phyllitic schist boulders (cf. text; 30 March 2008). (i and j) Lateral moraine system at Strauchon Glacier (21 February 2008, 7 December 2010; cf. Winkler, 2009a). (k) Whymper Glacier with its almost entirely debris-covered lower tongue (8 December 2009). (l) Lateral moraine (framed) in the outer western foreland of Whymper Glacier (27 March 2007; all photos: S. Winkler).

At Balfour Glacier, all sites were located in the lateral moraine system of the southern foreland. TCND samples were obtained from a section relatively high on the lateral moraine as the lower parts were inaccessible. Horace Walker Glacier is characterised by a comparatively large multi-ridged lateral moraine system, but dense vegetation and limited numbers of suitable boulders proved not suitable for Schmidt-hammer tests. A multi-ridge section of the northern lateral moraine at La Perouse Glacier was selected for sampling/testing, whereas relatively little research was carried out at Marchant Glacier because special local bedrock properties constituted a significant disadvantage (see section ‘Discussion’). By contrast, the multi-ridge lateral moraine complex at Strauchon Glacier with seven individual ridges above the local tree line constituted an ideal site as shown by the intense pilot study (cf. Winkler, 2009a). At Whymper Glacier, only a short single-ridge lateral moraine segment seemed suitable for fieldwork.

On all glacier forelands, the dominant bedrock is Haast Schist (Rakaia Terrane of the Torlesse Composite Terrane), subdivided into different textural/metamorphic zones becoming increasingly metamorphosed and deformed towards the Alpine Fault (Cox and Barrell, 2007). In a case study at Fox Glacier, Brook and Lucas (2012) have shown that different lithological properties of these zones exhibit an influence on clast shape. Although most glaciers studied flow like Fox Glacier through more than one subzone on their down-valley courses, the actual Schmidt-hammer sampling sites all belong to either textural zone III (schist) or IIB (semi-schist). There is a clear predominance of semi-schist boulders (zones IIB and IIA) at most sites dumped as supraglacial debris during lateral moraine formation and originating from the valley slopes up-glacier. Boulders of different bedrock, for example, those originating from the glaciers’ headwalls in non-schistose quartzo feldspathic sandstone, were avoided whenever possible. Hence, the structural and textural heterogeneity of the Haast Schist initially displayed no fundamental obstacle for the attempt to calculate a regional SHD age-calibration curve. All schists are described as well or highly foliated (Cox and Barrell, 2007), but mainly appear fairly massive and suitable for Schmidt-hammer testing despite the presence of quartz veins up to several centimetres thick. Those could easily be avoided with the Schmidt-hammer, but ensured sufficient quartz content of the samples taken for TCND.

Methods

Schmidt-hammer exposure-age dating

Developed as an instrument for in situ destruction-free testing of concrete hardness in construction works, the Schmidt-hammer has subsequently been introduced and applied in various fields of geomorphology and geology (Betts and Latta, 2000; Day and Goudie, 1977; Engel, 2007; Goudie, 2006; Haupt, 2012; McCarroll, 1987, 1994; Matthews and Winkler, 2011; Niedzielski et al., 2009; Shakesby et al., 2006, 2011; Viles et al., 2011). In the context of investigating Late Glacial and Holocene glacier chronologies, the Schmidt-hammer has been widely used as a relative-age dating technique (e.g. Aa and Sjåstad, 2000; Evans et al., 1999; McCarroll, 1989a, 1989b, 1991; Matthews and Shakesby, 1984; Winkler, 2000; Winkler and Shakesby, 1995). Such applications have successfully separated moraines formed during different glacier advance periods (LIA-type events sensu Matthews and Briffa, 2005). Pilot studies combined Schmidt-hammer measurements with available radiocarbon ages in order to achieve age constraints (e.g. Aa et al., 2007; Nesje et al., 1994; Winkler, 2005). SHD combining Schmidt-hammer and TCND has recently been successfully applied (e.g. Matthews and Owen, 2010; Matthews and Winkler, 2011; Shakesby et al., 2006). Schmidt-hammer tests have also been used to ensure the representativeness of boulders selected for TCND sampling (Putnam et al., 2012; Winkler, 2009a). The Schmidt-hammer N-type used here has a spring-suspension driven impact plunger calibrated to 2.207 Nm impact energy (Proceq, 2004). The rebound (R-)value obtained is determined by the compressive strength and hardness of the surface (see Adnan, 2009; Aydin and Basu, 2005; Christaras, 1996, for details).

The test design for field measurements followed previous studies (Winkler, 2005, 2009a) to ensure comparability and mimic successful adoption to local/regional conditions. Sample size (n) was 50 boulders tested with one impact each. Where possible, several test sites were allocated to one moraine ridge. Although there are different approaches regarding both sample size and impacts per boulder (cf. Haupt, 2012), the number of suitable boulders on moraine ridges did not allow more intense testing following, for example, Matthews and Owen (2010) or Shakesby et al. (2011). Otherwise, the common practices and recommendations on the choice of individual boulders as described in the literature were followed (McCarroll, 1994; Shakesby et al., 2006).

R-value raw data were subject to standard statistical treatment such as Kolmogorov–Smirnov tests (Sachs, 1999; Schönwiese, 1992) to ensure consistency of individual test sites. Additionally, visual examination of histograms was used to detect skewed and/or bimodal modal distributions caused, for example, by incorporation of previously weathered boulders or postdepositional erosion (cf. Appendix 1). Mean R-value and the standard error of the mean (SEM) at 95% confidence (α = 0.05) were calculated for every test site (Shakesby et al., 2006). Based on the empirical assumption that moraines formed during a LIA-type event should reveal overlapping SEM-intervals, a preliminary grouping of individual ridges was proposed. This was subsequently tested by applying Kruskal–Wallis analysis of variance (ANOVA) tests (cf. Lehmann, 2002; Sachs, 1999; Schönwiese, 1992) to reveal statistically significant age differences and confirm the initial grouping (Winkler, 2005, 2009a).

Terrestrial cosmogenic nuclide (¹⁰Be) exposure-age dating

Numerical (‘absolute’) age dating using terrestrial (in situ) produced cosmogenic nuclides has seen a significant progress over the past two decades (cf. Balco et al., 2008; Cockburn and Summerfield, 2004; Dunai, 2010; Fabel and Harbor, 1999; Gosse and Phillips, 2001; Watchman and Twidale, 2002). TCND has proven to be a useful supplement and/or alternative to radiocarbon (¹⁴C) dating. Apart from few noticeable exceptions (e.g. Jomelli et al., 2011; Kaplan et al., 2010; Putnam et al., 2012; Schaefer et al., 2009), financial, laboratory, and conservational constraints often vigorously limit the number of sampled boulders for TCND. The selection of the ‘right’ boulders, that is, those being representative for the moraine ridge, remains crucial to avoid potential chronological and palaeoclimatic misinterpretations. With laboratory precision and methodological procedures of TCND successfully improved in recent years, uncertainties arising from sampling ‘errors’ seem currently one of the greatest challenges of this dating technique (cf. Kirkbride and Winkler, 2012; Winkler and Matthews, 2010a).

To allow calibration with subsequent TCND age calculation, several parameters, including position and shielding, were directly measured in the field (cf. Appendix 2). Rock samples were knocked off manually, preferably from schist boulders with quartz veins. Sampling was restricted to the uppermost 5 cm or less on the boulder surfaces. Rock density for all samples was uniformly estimated to 2.65 g/cm³ for subsequent calculations. No micro-relief pattern indicating selective weathering and erosion (cf. Owen et al., 2007) was observed during sampling. Because these supposedly young surfaces did not allow any meaningful estimate of long-term surface weathering, an estimation of 1 cm surface weathering in 10,000 years derived from Late Glacial semi-schist boulders in the lower Tasman Valley was applied (S. Winkler, unpublished data). Most sampling sites were at altitudes of around 1000 m a.s.l., and under present climatic conditions, they are located around the lower limit of the average winter snow line (Fitzharris et al., 1992b), and annual snow cover will hardly add up to more than 1 month. The sampled boulders often ‘protrude’ above the moraine crests and will become snow-free at first. Consequently, no calibration for snow cover (cf. Reuther, 2007; Schildgen et al., 2005; Shakesby et al., 2008) was considered. This follows recent regional studies by Schaefer et al. (2009) and Putnam et al. (2012).

All TCNDs were prepared at the Cosmogenic Isotope Laboratory of the Department of Geological Sciences, University of Canterbury (New Zealand) following standard procedures (cf. Child et al., 2000; Mifsud et al., 2012; Putnam et al., 2012; Schaefer et al., 2009). Accelerator mass spectrometry (AMS)-measurements were performed at the Rafter Laboratory of the National Institute of Geological and Nuclear Sciences (GNS), Lower Hutt, New Zealand. Final age calculation and calibration was achieved with the CRONOS-Earth ¹⁰Be exposure-age calculator in its newest version 2.2 (cf. Balco et al., 2008) using the latest regional ¹⁰Be production rate for the Southern Alps (Putnam et al., 2010b). Because the resulting ages are considerably older than those obtained from the application of the ‘standard’ global production rate (10% and more), samples previously published (Winkler, 2009a) needed to be revised. Unlike the ¹⁰Be production rate chosen, age calibration with constant production rate models versus time-varying production models of the scaling scheme for spallation yielded only insignificant differences (in the order only tens of years or less). No special attention has consequently been paid to the discussion of those models (cf. Desilets and Zreda, 2003; Desilets et al., 2006; Dunai, 2000, 2001; Lal, 1991; Lifton et al., 2005; Stone, 2000). Because Schaefer et al. (2009) used the scaling scheme of Desilets, all final ages presented for discussion here apply the same scaling scheme to ensure compatibility.

There are several potential sources of error when calculating TCND ages for moraines. Inherent ages due to incomplete ‘zeroing’ of boulder surfaces are not considered a major issue because sampling sites were located on lateral moraines predominately formed by dumping of supraglacial debris (cf. Benn and Evans, 2010; Winkler, 2009b; Winkler and Hagedorn, 1999). Furthermore, Schaefer et al. (2009) and Putnam et al. (2012) empirically confirmed that inheritance is not a major source of uncertainty in this region. By contrast, underestimates of TCND age caused by subsequent exhumation of boulders or postdepositional modification have recently received some attention (e.g. Akcar et al., 2011; Applegate et al., 2012; Heyman et al., 2011; Stroeven et al., 2011). In this study, the actual position of most boulders selected from the crest of lateral moraine ridges made possible exhumation of deeply embedded boulders unlikely. Postdepositional disturbance connected to the high seismic activity of the nearby Alpine Fault or paraglacial erosion could, however, have affected the boulders sampled. This ‘geomorphic’ uncertainty was minimised by Schmidt-hammer tests on all boulders prior to TCND sampling (cf. Winkler, 2009a). In response to the discussion about the interpretation of multiple TCND ages obtained for individual moraine ridges (Applegate et al., 2010, 2012; Heyman et al., 2011; Stroeven et al., 2011), all moraine age estimations in this study are not a priori calculated as statistical means (cf. Putnam et al., 2012; Schaefer et al., 2009) but based on site-specific interpretation (see sections ‘Results’ and ‘Discussion’).

SHD age-calibration curves

The construction of SHD age-calibration curves to calibrate the Schmidt-hammer data using TCND followed the same steps successfully applied in a pilot study (see Winkler, 2009a, for details). A linear relationship between R-values and TCND ages was applied following comparable attempts for Holocene timescales (Matthews and McEwen, 2013; Matthews and Owen, 2010; Matthews and Winkler, 2011; Shakesby et al., 2011) and initial tests with other (e.g. power-law) equations which yielded lower coefficients of determination (R²) than linear ones. It was considered that moraine groups rather than individual moraine ridges should be dated. An age difference of at least 100–150 years between individual ridges formed during one LIA-type event can be expected, as confirmed by the LIA-moraine chronology in New Zealand and elsewhere (Davis et al., 2009; Grove, 2004; Winkler, 2004; Winkler and Matthews, 2010). Additionally, realistic approaches to the dating of moraines with the Schmidt-hammer (but also TCND) need to consider that error margins of individual samples (not subsequently calculated arithmetic means) do not allow for such a detailed differentiation. Consequently, following this consideration, R-value means for the moraine groups have been calculated and plotted against TCND ages as fixed points. However, rather than only calculating arithmetic means of all available TCND samples for these moraine groups (the approach used, for example, by Schaefer et al., 2009), single TCND ages have been excluded from the calculation of the means in order to evaluate their representativeness and detect potential outliers. In every case, several TCND ages were available for one moraine group: thus, alternative mean TCND ages were applied as fixed points in age-calibration curve construction, and the resulting R² values were compared. Finally, those curves with highest R² were considered as most reliable, and the interpretation of potential outliers among the TCND and Schmidt-hammer samples were based on these results.

Results

TCND and Schmidt-hammer data

All 29 TCND ages obtained (i.e. including potential outliers) are plotted on Figure 3 (see Table 2). Although errors of several samples are quite substantial, some clusters within individual forelands as well as overall can be recognised. However, the unequal spread of samples over the investigated forelands and their possible variable reliability means that the validity of these clusters must not be over-interpreted. Only a single TCND age from the crest of the innermost ridge of prominent moraine complex at Horace Walker Glacier was obtained (see above). Schmidt-hammer R-values at Marchant Glacier were much lower and not comparable to any other foreland. Even the innermost (recent) moraine sampled yielded extremely low R-value means, whereas a single TCND age confirmed the ‘recent’ (19th or early 20th century) age. The reason for those low R-values is the locally strongly foliated biotite greenschist (see Cox and Barrell, 2007) with its undulating boulder surfaces and highly variable point load strength depending on its direction to foliation (Augustinus, 1992). Furthermore, Marchant Glacier is entirely located in schist zone IV, and the semi-schist dominating at other sites is absent. Schmidt-hammer samples on two older latero-frontal moraines in the southern foreland show marginally insignificantly lower R-values (Tables 2 and 3). Albeit large error margins, TCND ages support the age-trend and are interpreted as (early) LIA-moraines. Overtopping of an older moraine can be ruled out by the unimodal R-value distribution and positive skewness.

Figure 3.

Visual comparison of all TCND ages (including their external errors) obtained from the six glacier forelands investigated (cf. Table 2).

Table 2.

TCND ages and Schmidt-hammer data obtained from the six glaciers. The surface exposure age was derived applying the regional ¹⁰Be production rate by Putnam et al. (2010b) and the scaling scheme of Desilets et al. (2006; cf. text and Appendix). External errors (following Balco et al., 2008) are added.

Site	TCND sample	¹⁰Be exposure age (years)	SH samples^a (n)	SH results^b (R-value)
Balfour Glacier
LM a	SK 43	2828 ± 665	5	55.08 ± 2.94; 54.58 ± 2.97; 54.38 ± 2.83; 53.32 ± 2.60; 52.66 ± 2.30
	SK 59	354 ± 295
	SK 60	497 ± 248
Horace Walker Glacier
LM a.4^c	SK 40	1482 ± 582	–	–
La Perouse Glacier
LFM d,1,2 (S)	–	–	8	61.60 ± 1.57; 62.48 ± 1.55; 61.30 ± 1.40; 62.04 ± 1,56; 61.84 ± 1,69; 61.86 ± 1.74; 61.02 ± 1.35; 61.08 ± 1.23
LM a (N)	SK 41	1468 ± 430	4	56.88 ± 2.05; 55.72 ± 1.92; 54.42 ± 2.18, 55.28 ± 2.32
LM a (N)	SK 58	1082 ± 314	4	56.88 ± 2.05; 55.72 ± 1.92; 54.42 ± 2.18, 55.28 ± 2.32
LM b (N)	SK 42	144 ± 418	1	56.44 ± 3.30
	SK 56	789 ± 258
	SK 57	873 ± 176
LM 1 (N)	–	–	4	60.62 ± 1.47; 58.10 ± 1.71; 58.70 ± 1.99; 59.56 ± 1.96
LM 2 (N)	–	–	2	60.18 ± 2.14; 59.96 ± 1.83
Marchant Glacier
LM a	SK 54	348 ± 221	1	26.60 ± 1.96
LM b	SK 53	545 ± 287	1	26.72 ± 2.23
LM 1	SK 55	203 ± 141	2	27.68 ± 2.14; 28.84 ± 2.16
Strauchon Glacier
LM a	SK 12	2809 ± 326	10	44.68 ± 2.37; 43.22 ± 2.20; 43.84 ± 1,84; 42.22 ± 2.37; 43.76 ± 2.25; 45.42 ± 2.10; 44.06 ± 2.33; 43.74 ± 1.67; 43.68 ± 2.41; 44.86 ± 2,01
	SK 48	2775 ± 265
	SK 49	2688 ± 278
LM b	SK 11	2710 ± 479	6	40.26 ± 2,50; 41.90 ± 2.28; 41.52 ± 2.06; 42.92 ± 2.36; 41.24 ± 2.29; 42.20 ± 2.34
	SK 36	2911 ± 402
	SK 37	2905 ± 370
	SK 37a	3054 ± 433
	SK 50	1932 ± 188
LM c	SK 13	2603 ± 296	2	41.56 ± 1.93; 41.12 ± 2.30
LM d	SK 35	1847 ± 416	5	46.46 ± 2.40; 45.98 ± 2.11; 45.40 ± 2.26; 46.44 ± 2.16; 45.32 ± 2.06
LM d	SK 35a	1914 ± 407	5
LM e	SK 10	1917 ± 259	9	47.68 ± 1,66; 45.42 ± 1,68; 46.04 ±± 1.96; 45.42 ± 1.91; 45.70 ± 2.02; 46.18 ± 2.27; 48.06 ± 2.40; 47.50 ± 1.84; 46.90 ± 2.28
LM e	SK 34	2228 ± 450	9
LM f	SK 51	1102 ± 241	13	48.96 ± 1.90; 49.42 ± 1.71; 49.00 ± 1.65; 47.72 ± 1.80; 48.86 ± 2.41; 48.96 ± 1.97; 48.92 ± 1.84; 49.72 ± 2.40; 47.56 ± 2.34; 49.92 ± 2.19; 48.68 ± 2.23; 48.24 ± 2.36; 48.50 ± 2.88
LM g	SK 38	992 ± 355	9	50.36 ± 1.53; 50.92 ± 1.58; 50.04 ± 2.58; 51.78 ± 2.34; 52.26 ± 1.56; 50.36 ± 2.55; 51.00 ± 2.01; 50.42 ± 2.68; 49.32 ± 2.53
Whymper Glacier
LM a	SK 8	1598 ± 574	3	52.38 ± 1.84; 50.28 ± 2.00; 50.10 ± 2.07
LM a	SK 9	327 ± 571	3	52.38 ± 1.84; 50.28 ± 2.00; 50.10 ± 2.07

Number of SH samples (each individual n = 50).

Mean R-values with SEM at 95% confidence (α = 0.05).

Innermost ridge of the complex latero-/latero-frontal moraine system (cf. text).

Table 3.

Results of two-sided Kruskal–Wallis ANOVA H-tests at α = 0.05 (with α – level of significance and adjusted for ties; cf. Sachs, 1999; Schönwiese, 1992) and H₀ (null hypothesis): Distribution of R-values is the same across all samples (calculated using IBM SPSS Statistics 19 software package).

Site(s)	φ ^a	Total n^b	H*^c	Decision
Balfour Glacier
LM a	4	250	0.653	Retain H₀
La Perouse Glacier
LFM d + 1 + 2 (S)	7	400	0.712	Retain H₀
LM 1 + 2 (N)	5	300	0.346	Retain H₀
LM a + b (N)	4	250	0.115	Retain H₀
LFM d/1/2 (S) + LM 1/2 (N)	13	700	0.030	Reject H₀
LFM d/1/2 (S) + LM 1/2 (N)	12^d	650	0.246	Retain H₀
LFM d/1/2 (S) + LM 2 (N)	10	550	0.484	Retain H₀
Marchant Glacier
LM a + b + 1	3	200	0.314	Retain H₀
Strauchon Glacier
LM a	9	500	0.815	Retain H₀
LM b	5	300	0.618	Retain H₀
LM c	1	100	0.740	Retain H₀
LM d	4	250	0.913	Retain H₀
LM e	8	450	0.507	Retain H₀
LM f	12	650	0.910	Retain H₀
LM g	8	450	0.808	Retain H₀
LM a + b	15	800	0.085	Retain H₀
LM a + b	14	750^d	0.259	Retain H₀
LM a + b + c	17	900	0.029	Reject H₀
LM a + b + c	16	850^d	0.079	Retain H₀
LM d + e	13	700	0.840	Retain H₀
LM f + g	21	1100	0.117	Retain H₀
LM a + b + c + d	21	1100	0.000	Reject H₀
LM a + b + c + d + e	30	1550	0.000	Reject H₀
LM d + e + f + g	35	1800	0.000	Reject H₀
LM d + e + f	26	1350	0.008	Reject H₀
LM e + f + g	30	1550	0.000	Reject H₀

Degrees of freedom (φ = k − 1; k = number of Schmidt-hammer samples).

Total of individual R-values/test impacts.

Asymptotic significance at α = 0.05 (cf. decision regarding H₀).

One sample (potential outliner) not included (cf. Table 2 and text).

The lateral moraine in the western foreland of Whymper Glacier gave rather variable ages, but Schmidt-hammer samples were fairly consistent. The young age of TCND sample SK 9 could well indicate postdepositional disturbance. The southern lateral moraine at Balfour Glacier was sampled in its up-valley section. Although not considered as ideal due to a large debris fan building up on its distal side, a single TCND age of c. 2830 years (SK 43) was promising and led to the re-sampling of the site. The additional ages are, however, much younger, making interpretation difficult. Schmidt-hammer data are consistent regarding the means, but most samples display large SEMs, high negative kurtosis, negative skewness and detectable bimodal trends. Those signs are indicative of ‘disturbed’ sites, that is, moraines that have either undergone significant postdepositional disturbance or partial overtopping by subsequent glacier advances. Given absence of any proximal ridges, the younger TCND ages may well represent overtopping during a LIA expansion of Balfour Glacier (see section ‘Discussion’).

In the south-western part of the La Perouse Glacier foreland, the three innermost latero-frontal moraine ridges can be linked to the LIA (and more recent advances). Morphologically, the two innermost ridges (LM 1 and 2) can be dated as ‘recent’, and although LFM d has higher developed vegetation cover, Schmidt-hammer data allow the ruling out of a pre-LIA age. An up-valley section of the northern lateral moraine complex yielded less conclusive data. The lower lateral moraine ridges at this site (LM 1/2) can be morphologically correlated with the southern latero-frontal moraines. Statistical tests, including all Schmidt-hammer samples from LM 1/2, initially fail, but if one individual sample (a potential outlier?) is excluded, the hypothesis can be retained (Table 3). There is apparently an R-value mean difference of 2 points between the southern (61.65) and northern foreland (59.52). Because small differences in this order of magnitude can occur within a single glacier foreland as a result of lithological variations or variable debris sources (cf. Winkler, 2005), all inner moraine ridges are interpreted as LIA-/recent moraines.

On the basis of the Schmidt-hammer data and two other TCND samples, SK 42 can be classified as outlier. The single Schmidt-hammer sample from LM b has an unusual high SEM, a strong negative skewness and a very wide spread of individual R-value readings (25–72). A small section of LM b also seems to have slumped and exhibits a double-crested morphology. As a consequence, all data from this moraine have to be interpreted with some caution as both, overtopping of an existing moraine as well as postdepositional disturbance (or rock fall from the steep valley sides) cannot definitely be ruled out. Although both TCND samples from the outer moraine are treated as reliable and Schmidt-hammer data are consistent, they display a considerable age contrast. As further TCND sampling was not possible, no final conclusion can be reached as to which of the two samples is the outlier (but see below). An individual Schmidt-hammer sample from LM a (46.84 ± 3.29, bimodal distribution, strong negative kurtosis) was not included in the summarised data set because of its high deviation from the other samples. It could, however, indicate that LM a has an ‘old’ precursor that had been overtopped by a subsequent advance represented by the other Schmidt-hammer samples and the two TCND ages (see section ‘Discussion’).

SHD age-calibration curves

The lateral moraine system on the western foreland of Strauchon Glacier had previously been chosen as key test, and after preliminary results were successfully obtained (see Winkler, 2009a, for details), the site was revisited for both additional TCND samples and Schmidt-hammer tests. The same approach to calculating the age-calibration curve as outlined in Winkler (2009a) was used (see section ‘Methods’). The revised version differs significantly regarding the TCND age estimates due to the now available regional ¹⁰Be production rate (Putnam et al., 2010b). The same grouping of moraine ridges was confirmed and retained (Tables 3 and 4). Owing to the overall limited number of samples (except for LM a/b/c), a rather simple approach for the detection of potential outliers based on recent work by Applegate et al. (2010, 2012) has been chosen. For moraine group LM a/b/c, the arithmetic mean is almost equivalent to the median of all samples if two samples (SK 50, 37a) are excluded. This coincides with the ‘normal distribution’ type of Applegate et al. (2010) and suggests that neither inheritance nor degradation constitutes a problem. SK 50 clearly is a young outlier, especially as the sample was not obtained from the morphologically innermost (youngest) moraine ridge. Despite some age difference between SK 37 and SK 37a (back-up sample from the same boulder), this most likely reflects a slightly different exposure history, and SK 37a was in the end not treated as an outlier (see below). With the remaining seven ages in close agreement, the arithmetic mean is considered more appropriate as the maximum age for this group of moraines and consequently adopted. Within moraine group LM d/e, sample SK 34 is considered an outlier and was therefore excluded, whereas both samples for moraine group LM f + g are used for an arithmetic mean. All decisions on outliers were checked by initial test runs of age-calibration curve calculation, including those samples with mean calculation which yielded considerably lower R² values (see section ‘Methods’). Input R-values for moraines and moraine groups are mean values based on all available samples. One exception is LM b where due to one comparably low SH test sample, two alternative R-value means each for the moraine and all related moraine groups (including and excluding this sample) have been used with the calculations (cf. Table 4b).

Table 4(a).

TCND and (b) Schmidt-hammer means used for initial calculation of alternative SHD age-calibrations curves for Strauchon Glacier.

Strauchon Glacier	n	Years ago mean	σ ^a	Median
LM a + b + c	9	2719	329	2775
LM a + b + c²	8	2817	156	2792
LM a + b + c³	7	2783	133	2775
LM d + e	4	1977	171	1916
LM d + e⁴	3	1893	40	1914
LM f + g	2	1047	78	1047

Table 4(b)

Strauchon Glacier	SH samples (n)	R-values (mean ± SEM)	σ
LM a	10	43.91 ± 0.57	0.91
LM b	6	41.67 ±± 0.72	0.90
LM b	5^e	41.96 ± 0.57	0.65
LM c	2	41.34 ± 0.43	0.31
LM b + c	8	41.59 ± 0.55	0.79
LM b + c	7^e	41.93 ± 0.42	0.56
LM a + b + c	18	42.88 ± 0.67	1.45
LM a + b + c	17^e	43.10 ± 0.60	1.26
LM d	5	45.92 ± 0.48	0.55
LM e	9	46.78 ± 0.66	1.01
LM d + e	14	46.47 ± 0.50	0.95
LM f	13	48.80 ± 0.38	0.69
LM g	9	50.72 ± 0.58	0.89
LM f + g	22	49.59 ± 0.51	1.23

Standard deviation of means.

Excluding SK 50 (cf. text).

Excluding SK 37a and SK 50 (cf. text).

Excluding SK 34.

Excluding one SH sample (cf. Table 3).

Several of the TCND age and R-values means tested as alternatives for the age-calibration curves at Strauchon Glacier show very high R² (Table 5). As its inclusion yielded the best-fit curve, the decision not to interpret SK 37a as an outlier seems justified. The selection of specific TCND ages or R-value means does not significantly influence the final curve. Even the inclusion of those TCND samples initially considered as outliers (see above) gave reasonable results. This is interpreted a sign of robustness of the chosen attempt. Finally, the equation with the highest R² has been selected (curve 5 on Table 5), with σ (TCND age means) and SEM (R-value means) as indicated error bars (Figure 4a). The TCND ages from La Perouse Glacier have subsequently been tested using the similar attempt. Surprisingly, both TCND samples for the outermost lateral moraine (LM a) produced equations with an acceptable R², but the younger age showed clearly a better fit (Table 5 and Figure 4b). For LM b, the younger of both remaining ages after exclusion of the obvious outlier (SK 42, see above) yielded a slightly higher R² than their mean.

Table 5.

Results (only best fit shown) of SHD age-calibration curves and applied raw data for La Perouse and Strauchon Glaciers (cf. Tables 2 and 4 and text).

TCND ages^a	SH R-value means^a	Equation^b	R ²
La Perouse Glacier
(1) 1468, 832, 1503^c	55.14, 56.44, 59.52	y = −0.0033x + 59.756	0.9557
(2) 1468, 873, 150	55.14, 56.44, 59.52	y = −0.0034x + 59.819	0.9696
(3) 1468, 789, 150	55.14, 56.44, 59.52	y = −0.0033x + 59.688	0.9398
(4) 1082, 832, 150	55.14, 56.44, 59.52	y = −0.0047x + 60.240	0.9991
(5) 1082, 873, 150	55.14, 56.44, 59.52	y = −0.0046x + 60.251	0.9941
(6) 1082, 789, 150	55.14, 56.44, 59.52	y = −0.0047x + 60.212	0.9996
Strauchon Glacier
(1) 2783, 1893, 1047	43.10, 46.47, 49.59	y = −0.0037x + 53.519	0.9999
(2) 2783, 1893, 1047	42.88, 46.47, 49.59	y = −0.0039x + 53.690	0.9993
(3) 2783, 1893, 1047	41.93, 46.47, 49.59	y = −0.0044x + 54.426	0.9916
(4) 2783, 1893, 1047	41.59, 46.47, 49.59	y = −0.0046x + 54.689	0.9816
(5) 2817, 1893, 1047	43.10, 46.47, 49.59	y = −0.0037x + 53.422	1
(6) 2817, 1893, 1047	42.88, 46.47, 49.59	y = −0.0038x + 53.592	0.9998
(7) 2817, 1893, 1047	41.93, 46.47, 49.59	y = −0.0043x + 54.319	0.9943
(8) 2817, 1893, 1047	41.59, 46.47, 49.59	y = −0.0045x + 54.779	0.9899
(9) 2719, 1893, 1047	43.10, 46.47, 49.59	y = −0.0039x + 53.707	0.9992
(10) 2817, 1977, 1047	43.10, 46.47, 49.59	y = −0.0037x + 53.515	0.9973

Different numbers are the result of inclusion/exclusion of single individual samples from TCND and SH R-value means (cf. text and Tables 2 and 4).

Equation for linear trend with related R².

Age set for recent/LIA moraines.

Figure 4.

(a) Best-fit SHD age-calibration curve for Strauchon Glacier (cf. Table 5, and text). (b) Best-fit SHD age-calibration curve for La Perouse Glacier. The best alternative curve (with SK 41 instead of SK 58) is indicated by dotted lines and symbols (cf. Table 5). (c) Comparison of the SHD age-calibration curves for La Perouse and Strauchon Glacier with TCND ages/Schmidt-hammer R-values from Balfour Glacier, Marchant Glacier and Whymper Glacier. Although exhibiting significantly lower R-values, an age-calibration curve for Marchant Glacier is likely to be roughly parallel. A single TCND age for Whymper Glacier falls within the range of the two above mentioned curves. Two alternative interpretations for the results from Balfour Glacier (cf. text) are also shown: one (alternative 1) falls far beyond these two curves, and the second (alternative 2) falls between them.

Discussion

Methodology

Although the application of SHD in general was successful and the age-calibration curve for Strauchon Glacier substantially improved, the attempt to develop a regionally applicable curve for the western part of the Southern Alps failed. Higher than expected lithological variation among the Haast Schist members was responsible for this failure. Differences in mean R-values for contemporaneous rock surfaces became apparent in the data from La Perouse Glacier and Strauchon Glacier. Although both are identically located regarding the different schist subzones that cross by the glacier foreland (Cox and Barrell, 2007), differences consistently range up to 5 points on the R-value scale, which are too large to support a single regional age-calibration curve. Variability of lithological properties and the composition of boulder samples at local scale, both apparently not sufficiently obvious during field sampling, can be accounted for these differences. However, individual curves exhibit parallel trends and slopes with comparable (linear) equations on all forelands, even at Marchant Glacier with its extremely low R-values (Figure 4c). Using these calibration curves, the age from Whymper Glacier (SK 8) falls between those for La Perouse Glacier and Strauchon Glacier. The relatively old TCND age from Balfour Glacier (SK 43) is related to R-values considerably higher than expected. Judging on field observations, it seems unlikely that lithology produces the much higher R-values at Balfour Glacier. Alternatively, it might be considered that this lateral moraine has subsequently been overtopped or influenced by postdepositional disturbance. Thus, the ‘old’ TCND age may well represent the true age of moraine formation, whereas the Schmidt-hammer data (some samples revealing a strong bimodal distribution) represent the latest episode of overtopping or disturbance. The two young TCND ages at first judged as young outliers (see above) are therefore considered indicative of a complex depositional history. The mean R-value is consistent with an LIA age for this overtopping event.

With the development of a regional SHD age-calibration curve prohibited by the lithological properties, more sites on individual forelands would have decreased error margins for the Schmidt-hammer data. Furthermore, following previous studies (Winkler, 2005, 2009a), linear equations were used for the construction of SHD age-calibration curves. Alternatively, Stahl et al. (2013) proposed a power-law equation as best-fitting option for SHD of late Quaternary fluvial terraces east of the Main Divide. As they cover a longer time period and deal with a different rock type, this does not necessarily have implications for this study. Unpublished data by the author suggest that for greywacke rocks east of the Main Divide semi-logarithmic equations give better results for tests, including early-Holocene and older moraines (Winkler, 2012). Several recent studies argue that a linear equation is the most appropriate type of SHD age-calibration curves for the Holocene (e.g. Matthews and McEwen, 2013; Matthews and Owen, 2010; Shakesby et al., 2011). Most of these studies were performed on highly weathering-resistant rocks with a slow decrease in R-values with time than is appropriate for the Haast Schist members. Consequently, extending linear equations shown here for the late Holocene might not be valid for the whole Holocene as the R-value decrease might near the mechanical resolution of the instrument and/or the time the exposed surfaces face widespread removal of detritus. But for the c. 3000 years covered in this study, linear equations clearly provide the best fit.

Unlike SHD, there is potential for substantial improvement of TCND. Due to limited financial resources, the focus of TCND was directed towards the glacier foreland of Strauchon Glacier, where the potential for useful results was greatest. This naturally left other forelands with low numbers of dates but with some potential if combined with SHD. Error bars of TCND ages presented here are considerably higher than those achieved by Schaefer et al. (2009) or Putnam et al. (2012), but represent the laboratory precision available to the author. Despite higher individual error bars, the spread of individual TCND ages for moraines and moraine groups at Strauchon Glacier is in the same order as for moraine ages reported by these authors. They should, therefore, not be regarded a priori as less reliable in the light of geomorphological uncertainties (Kirkbride and Winkler, 2012) and the natural spread of ages produced during a LIA-type event (Winkler and Matthews, 2010). Compared to the those TCND ages reported by Winkler (2009a) using the global ¹⁰Be production rate of Balco et al. (2008), the new regional production rate of Putnam et al. (2010b) results in considerably (>10%) older TCND ages. But as this is the only regional production rate available to date, its application was without alternative.

Glacier chronology and correlation

Previous chronological studies are available for some of the glaciers in form of radiocarbon ages published by Wardle (1973, 1978) and Röthlisberger (1986). Several radiocarbon ages have been obtained from the prominent latero-frontal moraine system at Horace Walker Glacier (cf. Röthlisberger, 1986). With just one TCND age for this glacier, the existing radiocarbon chronology cannot be used for comparison and a postulated advance at c. 1450 ¹⁴C yr BP (1300 cal. yr BP using SHCal 04 by McCormac et al., 2004) roughly corresponds to it. Röthlisberger (1986) claims – unlike Wardle (1978) – that the innermost moraine ridge is younger (800 cal. yr BP), but the raw data provided are not conclusive and, hence, both interpretations remain viable.

At Balfour Glacier, Röthlisberger (1986) does not report any major advance prior to c. 1300 cal. yr BP. A second advance with similar extent is reported for 570–530 cal. yr BP, an age that could correspond to one of the younger TCND ages obtained. But, except for this vague support for the overtopping hypothesis mentioned above, there is no obvious correspondence between the radiocarbon and the SHD/TCND chronologies at Balfour Glacier. An old advance at La Perouse Glacier around or prior to 2.850 cal. yr BP (Röthlisberger, 1986) lacks a similar TCND age as the older samples from that moraine (SK 41) could theoretically indicate a two-phase formation, but at a considerable younger date. Only one single Schmidt-hammer sample deviating from all other samples could indicate any older advance that remains, therefore, purely speculative. Wardle (1978) reports an age of c. 5260–5055 cal. yr BP from the outermost terminal moraine at La Perouse Glacier subsequently used by Porter (2000) as evidence for the onset of Neoglaciation in the western Southern Alps. No additional advances prior to the LIA have been reported for La Perouse Glacier, thus no correspondence to the other TCND ages can be established. Finally, the outer ridges of the prominent lateral moraine system at Strauchon Glacier should, according to Röthlisberger (1986), date from the Late Glacial. TCND ages and Schmidt-hammer data enable rejection of this hypothesis. Two pre-LIA advances during the late Holocene are mentioned by Röthlisberger (1986): one around 2150 cal. yr BP and the other around 1375 cal. yr BP. Both do not correspond with the detailed SHD chronology and give the impression that surface exposure dating methods (SHD, TCND) and radiocarbon dating do not correspond well in the specific study area.

Interesting comparisons can be made between the glaciers studied and the most recent TCND-derived glacier chronology for the Southern Alps. Putnam et al. (2012) amalgamated their data from Cameron Glacier (south-eastern Arrowsmith Range, some 70 km east of the study area) with their previous work at Hooker Glacier, Mueller Glacier and Tasman Glacier immediately east of the Main Divide near Mt Cook (Schaefer et al., 2009). On the basis of over 100 individual TCND ages, they conclude an asynchronous pattern of Holocene glacier variations in the Northern and Southern Hemisphere, mainly driven by latitudinal migration of the inter-tropical convergence zone (ITCZ) and related shifts within the atmospheric circulation pattern. There is no good correspondence between their TCND ages and those obtained in the present study, except from the ubiquitous LIA-signal and a vague overlap with moraines formed around ~2000 and ~1000 years ago (see Figure 5). Apart from the above mentioned old date for the outermost terminal moraine at La Perouse Glacier (Wardle, 1978) and the disputable Late Glacial/early-Holocene origin for the Horace Walker Glacier moraine complex (Röthlisberger, 1986), there is also a lack of evidence for early-Holocene or mid-Holocene glacier advances in the West. This discrepancy can be interpreted in two ways: either as an indication of spatial differentiation and diverging patters east and west of the Main Divide, or as a hint that the data presented here are unreliable. At least for Strauchon Glacier, the author sees, however, no reason to question the reliability of these data.

Figure 5.

Comparison of the comprehensive TCND chronology of Schaefer et al. (2009) and Putnam et al. (2012) for Cameron Glacier, Hooker Glacier, Mueller Glacier and Tasman Glacier with the results of this study. The latter are displayed as LIA-type events rather than ‘glacial pulses’ (see text). The ubiquitous LIA is not shown in detail here.

Even without considering the recently discussed potential influence of large mass movements upon moraine formation (Reznichenko et al., 2011, 2012; Shulmeister et al., 2009), the postulated representativeness of the TCND chronology of Putnam et al. (2012) may also be critically reassessed. The amalgamation of both data sets to reconstruct a Holocene glacier chronology for the entire Southern Alps seems questionable. Cameron Glacier is much smaller and located in a relatively leeward position to the prevailing westerly airflow. It is, furthermore, the only reported glacier in the Southern Alps with numerous dated early-Holocene moraines. Evidence of those early-Holocene advances is lacking at Hooker Glacier, Mueller Glacier or Tasman Glacier, other than the 6900-year advance vaguely correlating with the 6500/6400-year moraines at Tasman Glacier and Mueller Glacier, respectively (Schaefer et al., 2009). By contrast, none of the late-Holocene advances prior to early LIA is reported from Cameron Glacier, whereas several moraines between 3200 and 1000 years ago are dated at Mt Cook. There are, additionally, considerable differences among the three glaciers at Mt Cook. The main problem of this amalgamation seems, however, the well-documented complicated response of large debris-covered glaciers (Hooker Glacier, Mueller Glacier and Tasman Glacier) compared to the simpler response characteristics of glaciers like the Cameron Glacier (Kirkbride, 2000).

It has to be noted that Schaefer et al. (2009) and Putnam et al. (2012) apply a concept of ‘glacial pulses’ for their chronology rather than the concept of LIA-type events sensu Matthews and Briffa (2005), the latter widely accepted and used by most recent chronological studies (cf. Davis et al., 2009). They relate a dated moraine to an advance (‘glacial pulse’) of the individual glacier only and neither collate those advances with respect to greater events like the LIA nor consider different response times with the correlation of advances at those glaciers. By contrast, the concept of LIA-type events addresses the differences in individual response time of glaciers (empirically shown for the LIA). At Tasman Glacier, the last overtopping of the latero-terminal moraine representing the LIA maximum occurred c. 170–180 years later than the lichenometrically dated LIA maximum at Mueller Glacier (Gellatly, 1982, 1984; Winkler, 2000, 2004). As such, a considerable difference in response times to a major mass balance disturbance seems not unlikely considering the different sizes and dynamics of the glaciers. The 1800-year moraine of Mueller Glacier may well represent the same advance period as the 1650-year moraine at Tasman Glacier (and the 1850-year advance at Strauchon Glacier!). If the concept of LIA-type events and their more realistic error margins are applied to the chronology of Schaefer et al. (2009), the initial mismatch with data from the present study will be considerably downsized.

The ‘glacial pulse’ concept makes it relatively easy to dismiss synchronous glacier variations on basis of individual glaciers and single advances. Within the LIA, even glaciers within sub-regions do not necessarily correlate in this context (Grove, 2004). This is also characteristic of widespread LIA-type events in the Northern Hemisphere that partly are dated to periods within those glacier advances that have also been reported from the Southern Alps. For example, Ivy-Ochs et al. (2009) report widespread advances associated with Göschener I oscillations (3.0–2.3 kyr BP) in the European Alps, and Menounos et al. (2009) report widespread advances between 3.5 and 2.7 kyr BP in Western Canada. Bakke et al. (2008) concluded that an increase in precipitation led to ice growth in Southern Norway around 2.8 kyr BP, and in other Scandinavian regions, glaciers were at least as large as they are today around 3.0 kyr BP (Matthews and Dresser, 2008; Nesje, 2009). Holzhauser and Zumbühl (1999), Hormes et al. (2001, 2006), Joerin et al. (2006) and Nussbaumer et al. (2011) all present data that individual glaciers in the European Alps advanced around 2.8 kyr BP. These data correspond well to a signal of glacial activity from Strauchon Glacier and two others. Even the 3.2 kyr BP advance from Mueller Glacier (Schaefer et al., 2009) is not too far away. From 1.9 to 1.6 kyr BP, there is likely to have been advances in Scandinavia (Nesje, 2009). Again, contemporary advances are reported from Strauchon as well as from Mueller Glacier.

Although the New Zealand data base is still comparably incomplete and inferior to the chronologies available for the European Alps and Scandinavia, the data in this study do not support the conclusion of an asynchronous pattern between hemispheres. By contrast, there are clear signs of synchronous advance periods. It is, however, an undisputed fact that glaciers in the Southern Alps experienced their Neoglacial maximum prior to the LIA what represent an important difference to many glacier regions in the Northern Hemisphere. But the parallel patterns of glaciers in New Zealand and maritime Norway that occurred during the 20th century (Chinn et al., 2005) may not constitute a singularity as synchronous late-Holocene advance periods are still partly supported by the newly achieved age constraints.

Conclusion

The present study focussed on 6 glacier forelands located west of the Main Divide where a total of 29 TCND samples have been collected and extensive Schmidt-hammer measurements have allowed SHD with reduced geomorphological uncertainties, associated with the palaeoclimatic interpretation of moraines (cf. Kirkbride and Winkler, 2012). The following conclusions can be drawn from these investigations:

SHD was successfully applied at the local scale on a number of glacier forelands based on linear age-calibration curves, which gave the best fit for the late Holocene (i.e. the last 3000 years).

Lithological variability of the dominating Haast Schist members prohibited developing a regional age-calibration curve. However, the individual calibration curves display very similar slopes and are comparable. This allows at least a preliminary stage assessment of the representativeness of individual TCND ages on glacier forelands with few TCND samples.

Cluster of moraine ages point towards the occurrence of LIA-type events around 2800, 1850–1450 and 1100–800 years ago, followed by the LIA commencing c. 500 years ago. No indications of earlier Neoglacial advances were found.

There is no good agreement with previously published radiocarbon ages from the investigated glaciers. At first sight, there is also hardly any agreement with the extensive TCND chronology for the Southern Alps provided by Schaefer et al. (2009) and Putnam et al. (2012). This statement needs, however, to be partly revised if different approaches to the palaeoclimatological interpretation of moraine ages are taken into account. With LIA-type events rather than ‘glacial pulses’ considered, some agreement does emerge.

On basis of dates from this study, no definite conclusions can be reached regarding the synchronous or asynchronous behaviour of mid-latitudinal Holocene mountain glaciers. Whereas most glaciers in the Southern Alps (unlike those in Scandinavia and the European Alps) mainly experienced their Neoglacial maximum prior to the LIA, some of the glacier expansion episodes outlined here fall within established expansion periods in the Northern Hemisphere. More research is, therefore, indispensable.

Footnotes

Appendix

Appendix 2.

Raw input data for the CRONOS age calculator

The year stands for the sampling year (not necessarily the year the samples were processed; cf. Balco et al., 2008; see also text). Shielding factors have been calculated using the geometric shielding calculator as integrated in the 2.2 version of the CRONOS-Earth 10Be-26Al exposure-age calculator (Balco et al., 2008; http://www.hess.ess.washington.edu). As the samples have been processed in batches, different 10Be-standards have been used, and carrier/chemistry blank correction has been performed on an individual basis.

Sample	Site	Year	Latitude (south)	Longitude (east)	Altitude (m a.s.l.)	Shielding factor	Thickness (cm)	Density (g/cm³)	Erosion (cm/a)	¹⁰Be (atoms/g)	¹⁰Be standard
Balfour Glacier
SK 43	LM a	2007	−43.553	170.087	1110.0 std	0.9354	5	2.65	0.0001	25,136.935 ± 5868	07KNSTD
SK 59	LM a	2008	−43.554	170.082	1070.0 std	0.928	5	2.65	0.0001	3132.690 ± 2609	07KNSTD
SK 60	LM a	2008	−43.553	170.084	1080.0 std	0.9363	5	2.65	0.0001	4471.442 ± 2228	07KNSTD
Horace Walker Glacier
SK 40	LM a.4	2007	−43.679	170.911	1075.0 std	0.977	5	2.65	0.0001	13,551.659 ± 5307	07KNSTD
La Perouse Glacier
SK 41	LM a (N)	2007	−43.572	170.033	990.0 std	0.9599	5	2.65	0.0001	12,272.054 ± 3577	07KNSTD
SK 58	LM a (N)	2008	−43.572	170.029	1000.0 std	0.9376	5	2.65	0.0001	8973.611 ± 2597	07KNSTD
SK 42	LM b (N)	2007	−43.571	170.023	990.0 std	0.9564	5	2.65	0.0001	1166.514 ± 3376	07KNSTD
SK 56	LM b (N)	2008	−43.572	170.029	1000.0 std	0.9522	5	2.65	0.0001	6685.376 ± 2167	07KNSTD
SK 57	LM b (N)	2008	−43.572	170.03	1000.0 std	0.9522	5	2.65	0.0001	7380.417 ± 1476	07KNSTD
Marchant Glacier
SK 54	LM a	2008	−43.614	170.001	1195.0 std	0.93	5	2.65	0.0001	3421.690 ± 2177	07KNSTD
SK 53	LM b	2008	−43.613	170.001	1180.0 std	0.9445	5	2.65	0.0001	5364.756 ± 2826	07KNSTD
SK 55	LM 1	2008	−43.624	170.001	1150.0 std	0.927	5	2.65	0.0001	1920.620 ± 1332	07KNSTD
Strauchon Glacier
SK 12	LM a	2006	−43.623	170.023	1155.0 std	0.9581	5	2.65	0.0001	29,079.733 ± 3303	NIST 30600
SK 48	LM a	2008	−43.624	170.028	1140.0 std	0.9742	5	2.65	0.0001	26,333.595 ± 2444	07KNSTD
SK 49	LM a	2008	−43.623	170.023	1155.0 std	0.9724	5	2.65	0.0001	25,779.972 ± 2602	07KNSTD
SK 11	LM b	2006	−43.623	170.023	1160.0 std	0.9773	5	2.65	0.0001	28,725.877 ± 5028	NIST 30600
SK 36	LM b	2007	−43.622	170.023	1160.0 std	0.9614	5	2.65	0.0001	27,708.059 ± 3768	07KNSTD
SK 37	LM b	2007	−43.623	170.023	1160.0 std	0.9668	5	2.65	0.0001	27,805.957 ± 3479	07KNSTD
SK 37a	LM b	2007	−43.623	170.023	1160.0 std	0.9668	5	2.65	0.0001	29,300.000 ± 4105	07KNSTD
SK 50	LM b	2008	−43.624	170.023	1140.0 std	0.9709	5	2.65	0.0001	18,377.661 ± 1745	07KNSTD
SK 13	LM c	2006	−43.624	170.023	1145.0 std	0.9778	5	2.65	0.0001	27,272.044 ± 044	NIST 30600
SK 35	LM d	2007	−43.622	170.024	1145.0 std	0.967	5	2.65	0.0001	17,588.769 ± 3937	07KNSTD
SK 35a	LM d	2007	−43.622	170.024	1145.0 std	0.967	5	2.65	0.0001	18,239.529 ± 3861	07KNSTD
SK 10	LM e	2006	−43.624	170.024	1140.0 std	0.9694	5	2.65	0.0001	19,942.011 ± 2651	NIST 30600
SK 34	LM e	2007	−43.624	170.024	1115.0 std	0.9695	5	2.65	0.0001	20,688.089 ± 4137	07KNSTD
SK 51	LM f	2008	−43.624	170.024	1110.0 std	0.9706	5	2.65	0.0001	10,533.338 ± 2289	07KNSTD
SK 38	LM g	2007	−43.624	170.025	1100.0 std	0.9689	5	2.65	0.0001	9332.817 ± 3328	07KNSTD
Whymper Glacier
SK 8	LM a	2006	−43.448	170.368	1050.0 std	0.9585	5	2.65	0.0001	15,296.746 ± 5474	NIST 30600
SK 9	LM a	2006	−43.44	170.371	980.0 std	0.9533	5	2.65	0.0001	3010.524 ± 5258	NIST 30600

Acknowledgements

Research and on-site landing permissions were granted by the Department of Conservation’s Westland Regional Office. Christina Wachler, Nina Kurr (University of Würzburg) and Yani Najman (Lancaster University) assisted during fieldwork. Sample preparation at UC was done by Rob Spiers, Sacha Baldwin-Cunningham and James Oram, and AMS-measurements at GNS by Albert Zondervan. The author wants to express his thanks for their great job. The manuscript benefitted from the valuable comments of four anonymous reviewers.

Funding

Fieldwork in New Zealand and TCND dating were financed by the Deutsche Forschungsgemeinschaft (DFG-contracts WI 1701/3 and 1701/4).

References

Sjåstad

(2000) Schmidt hammer age evaluation of the moraine sequence in front of Bøyabreen, western Norway. Norsk Geologisk Tidsskrift 80: 27–32.

Sjåstad

Sønstegaard

. (2007) Chronology of Holocene rock-avalanche deposits based on Schmidt-hammer relative dating and dust stratigraphy in nearby bog deposits, Vora, inner Nordfjord, Norway. The Holocene 17: 955–964.

Adnan

(2009) ISRM suggested method for determination of the Schmidt hammer rebound hardness revised version. International Journal of Rock Mechanics and Mining Sciences 46: 627–634.

Akcar

Ivy-Ochs

Kubik

. (2011) Post-depositional impacts on ‘Findlinge’ (erratic boulders) and their implications for surface-exposure dating. Swiss Journal of Geosciences 104: 445–453.

Anderson

Mackintosh

(2006) Temperature change is the major driver of late-glacial and Holocene glacier fluctuations in New Zealand. Geology 34: 121–124.

Anderson

Mackintosh

(2012) Controls on mass balance sensitivity of maritime glaciers in the Southern Alps, New Zealand: The role of debris cover. Journal of Geophysical Research 117: F01003.

Anderson

Lawson

Owens

(2008) Response of Franz Josef Glacier Ka Roimata o Hine Hukatere to climate change. Global and Planetary Change 63: 23–30.

Applegate

Urban

Keller

. (2010) Modeling the statistical distributions of cosmogenic exposure dates from moraines. Geoscientific Model Development 3: 293–307.

Applegate

Urban

Keller

. (2012) Improved moraine age interpretations through explicit matching of geomorphologic process models to cosmogenic nuclide measurements from single landforms. Quaternary Research 77: 293–304.

10.

Augustinus

(1992) The influence of rock mass strength on glacial valley cross-profile morphometry: A case study from the Southern Alps, New Zealand. Earth Surface Processes and Landforms 17: 39–51.

11.

Aydin

Basu

(2005) The Schmidt hammer in rock characterization. Engineering Geology 81: 1–14.

12.

Bakke

Lie

Dahl

. (2008) Strength and spatial patterns of the Holocene wintertime westerlies in the NE Atlantic region. Global and Planetary Change 60: 28–41.

13.

Balco

Stone

Lifton

. (2008) A complete and easily accessible means of calculating surface exposure ages or erosion rates from ¹⁰Be and ²⁶Al measurements. Quaternary Geochronology 3: 174–195.

14.

Barrell

DJA

Andersen

Denton

(2011) Glacial Geomorphology of the Central South Island (GNS Science Monographs 27). Lower Hutt: GMS Science, 81 pp.

15.

Benn

Evans

DJA

(2010) Glaciers and Glaciation. 2nd Edition. London: Hodder, 802 pp.

16.

Betts

Latta

(2000) Rock surface hardness as an indicator of exposure age an archaeological application of the Schmidt hammer. Archaeometry 42: 209–223.

17.

Böhlert

Egli

Maisch

. (2011) Application of a combination of dating techniques to reconstruct the Lateglacial and early Holocene landscape history of the Albula region (eastern Switzerland). Geomorphology 127: 1–13.

18.

Brook

Lukas

(2012) A revised approach to discriminating sediment transport histories in glacigenic sediments in a temperate alpine environment: A case study from Fox Glacier, New Zealand. Earth Surface Processes and Landforms 37: 895–900.

19.

Burrows

(1973) Studies on some glacial moraines in New Zealand – 2. Ages of moraines of the Mueller, Hooker and Tasman Glaciers. New Zealand Journal of Geology and Geophysics 16: 831–855.

20.

Burrows

(1980) Radiocarbon dates for post-Otiran glacial activity in the Mount Cook region, New Zealand. New Zealand Journal of Geology and Geophysics 23: 239–248.

21.

Burrows

(1989) Aranuian radiocarbon dates from moraines in the Mount Cook region, New Zealand. New Zealand Journal of Geology and Geophysics 32: 205–216.

22.

Burrows

Gellatly

(1982) Holocene glacier activity in New Zealand. Striae 18: 41–47.

23.

Burrows

Bell

Grant

(2002) Two new radiocarbon ages for mid- and late-Aranuian age valley-train deposits of the Franz Josef Glacier, Westland, New Zealand. Journal of the Royal Society of New Zealand 32: 415–425.

24.

Child

Elliott

Mifsud

. (2000) Sample processing for earth science studies at ANTARES. Nuclear Instruments and Methods in Physics Research Section B: Beam Interactions with Materials and Atoms 172: 856–860.

25.

Chinn

Fitzharris

Willsman

. (2012) Annual ice volume changes 1976–2008 for the New Zealand Southern Alps. Global and Planetary Change 92–93: 105–118.

26.

Chinn

TJH

(1996) New Zealand glacier responses to climate change of the past century. New Zealand Journal of Geology and Geophysics 39: 415–428.

27.

Chinn

TJH

(2001) Distribution of the glacial water resources of New Zealand. Journal of Hydrology (New Zealand) 40: 139–187.

28.

Chinn

TJH

Whitehouse

(1980) Glacier snow line variations in the Southern Alps, New Zealand. In: Proceedings of the Riederalp Workshop, September 1978, pp. 219–228, IAHS Publication No. 126. Oxfordshire: IASH.

29.

Chinn

TJH

Salinger

Fitzharris

. (2008) Glaciers and climate. Bulletin of the Federal Mountain Clubs of New Zealand, vol. 171, pp. 1–15.

30.

Chinn

TJH

Winkler

Salinger

. (2005) Recent glacier advances in Norway and New Zealand – A comparison for their glaciological and meteorological causes. Geografiska Annaler – Series A: Physical Geography 87: 141–157.

31.

Christaras

(1996) Non destructive methods for investigation of some mechanical properties of natural stones in the protection of monuments. Bulletin of the International Association of Engineering Geology 54: 59–63.

32.

Cockburn

HAP

Summerfield

(2004) Geomorphological applications of cosmogenic isotope analysis. Progress in Physical Geography 28: 1–42.

33.

Cox

Barrell

DJA

(2007) Geology of the Aoraki Area – Geological Map 1:250 000 (Qmap series 15). Lower Hutt: GNS Science, 71 pp.

34.

Davis

Menounos

Osborn

(2009) Holocene and latest Pleistocene alpine glacier fluctuations: A global perspective. Quaternary Science Reviews 28: 2021–2033.

35.

Day

Goudie

(1977) Field assessment of rock hardness using the Schmidt test hammer. Technical Bulletin: British Geomorphological Research Group 18: 19–29.

36.

Desilets

Zreda

(2003) Spatial and temporal distribution of secondary cosmic-ray nuclear intensities and applications to in situ cosmogenic dating. Earth and Planetary Science Letters 206: 21–42.

37.

Desilets

Zreda

Prabu

(2006) Extended scaling factors for in situ cosmogenic nuclides: New measurements at low latitude. Earth and Planetary Science Letters 246: 265–276.

38.

Dunai

(2000) Scaling factors for production rates of in situ produced cosmogenic nuclides: A critical reevaluation. Earth and Planetary Science Letters 176: 157–169.

39.

Dunai

(2001) Influence of secular variation of the geomagnetic field on production rates of in situ produced cosmogenic nuclides. Earth and Planetary Science Letters 193: 197–212.

40.

Dunai

(2010) Cosmogenic Nuclides: Principles, Concepts and Applications in Earth Surface Sciences. Cambridge: Cambridge University Press, 187 pp.

41.

Engel

(2007) Measurement and age assignment of intact rock strength in the Krkonoše Mountains, Czech Republic. Zeitschrift für Geomorphologie NF, Supplement 51: 69–80.

42.

Evans

DJA

Archer

Wilson

DJH

(1999) A comparison of the lichenometric and Schmidt hammer dating techniques based on data from the proglacial areas of some Icelandic glaciers. Quaternary Science Reviews 18: 13–41.

43.

Fabel

Harbor

(1999) The use of in-situ produced cosmogenic radionuclides in glaciology and glacial geomorphology. Annals of Glaciology 28: 103–110.

44.

Fitzharris

Hay

Jones

(1992a) Behaviour of New Zealand glaciers and atmospheric circulation changes over the past 130 years. The Holocene 2: 97–106.

45.

Fitzharris

Owens

Chinn

TJH

(1992b) Snow, ice and glacier hydrology.In: Mosley

(ed.) Waters of New Zealand. Wellington: The New Zealand Hydrological Society, pp. 75–93.

46.

Gellatly

(1982) Lichenometry as a relative-age dating method in Mount Cook National Park, New Zealand. New Zealand Journal of Botany 20: 343–353.

47.

Gellatly

(1984) The use of rock weathering-rind thickness to redate moraines in Mount Cook National Park, New Zealand. Arctic and Alpine Research 16: 225–232.

48.

Gellatly

(1985a) Glacial fluctuations in the central Southern Alps, New Zealand: Documentation and implications for environmental change during the last 1000 years. Zeitschrift für Gletscherkunde und Glazialgeologie 21: 259–264.

49.

Gellatly

(1985b) Historical records of glacier fluctuations in Mt Cook National Park, New Zealand: A century of change. The Geographical Journal 151: 86–99.

50.

Gellatly

Chinn

TJH

Röthlisberger

(1988) Holocene glacier variations in New Zealand: A review. Quaternary Science Reviews 7: 227–242.

51.

Gosse

Phillips

(2001) Terrestrial in situ cosmogenic nuclides: Theory and application. Quaternary Science Reviews 20: 1475–1560.

52.

Goudie

(2006) The Schmidt hammer in geomorphological research. Progress in Physical Geography 30: 703–718.

53.

Grove

(2004) Little Ice Ages, vol. 2. London: Routledge, 718 pp.

54.

Haupt

(2012) Anwendungsorientierte Bewertung zweier Methoden zur relativen Altersdatierung holozäner Moränen an Hochgebirgsgletschern. Unpublished Diploma Thesis, Universität Würzburg, Würzburg, 126 pp.

55.

Heyman

Stroeven

Harbor

. (2011) Too young or too old: Evaluating cosmogenic exposure dating based on an analysis of compiled boulder exposure ages. Earth and Planetary Science Letters 302: 71–80.

56.

Holzhauser

Zumbühl

(1999) Nacheiszeitliche Gletscherschwankungen (Blatt 3.8, Hydrologischer Atlas der Schweiz). Bern: Bundesamt für Landestopographie (Map).

57.

Hooker

Fitzharris

(1999) The correlation between climatic parameters and the retreat and advance of Franz Josef Glacier, New Zealand. Global and Planetary Change 22: 39–48.

58.

Hormes

Beer

Schlüchter

(2006) A geochronological approach to understanding the role of solar activity on Holocene glacier length variability in the Swiss Alps. Geografiska Annaler – Series A: Physical Geography 88: 281–294.

59.

Hormes

Müller

Schlüchter

(2001) The Alps with little ice evidence for eight Holocene phases of reduced glacier extent in the Central Swiss Alps. The Holocene 11: 255–265.

60.

IPCC (ed.) (2007) Climate Change 2007: The Physical Science Basis (Contribution of Working Group I to the 4th Assessment Report of the Intergovernmental Panel on Climate Change). Cambridge: Cambridge University Press, 996 pp.

61.

Ivy-Ochs

Kerschner

Maisch

. (2009) Latest Pleistocene and Holocene glacier variations in the European Alps. Quaternary Science Reviews 28: 2137–2149.

62.

Joerin

Stocker

Schlüchter

(2006) Multicentury glacier fluctuations in the Swiss Alps during the Holocene. The Holocene 16: 697–704.

63.

Jomelli

Khodri

Favier

. (2011) Irregular tropical glacier retreat over the Holocene epoch driven by progressive warming. Nature 474: 196–199.

64.

Kaplan

Schaefer

Denton

. (2010) Glacier retreat in New Zealand during the Younger Dryas stadial. Nature 467: 194–197.

65.

Kirkbride

(2000) Ice marginal geomorphology and Holocene expansion of debris-covered Tasman Glacier, New Zealand. In: Nakawo

Raymond

Fountain

(eds) Debris-Covered Glaciers (IAHS Publication No. 264). Paris: IAHS, pp. 211–217.

66.

Kirkbride

Winkler

(2012) Correlation of late Quaternary moraines: Impact of climate variability, glacier response, and chronological resolution. Quaternary Science Reviews 46: 1–29.

67.

Lal

(1991) Cosmic ray labeling of erosion surfaces: In situ nuclide production rates and erosion models. Earth and Planetary Science Letters 104: 424–439.

68.

Larsen

Davies

TRH

McSaveney

(2005) A possible coseismic landslide origin of late Holocene moraines of the Southern Alps, New Zealand. New Zealand Journal of Geology and Geophysics 48: 311–314.

69.

Lehmann

(2002) Statistik. Heidelberg: Spektrum, 385 pp.

70.

Lifton

Bieber

Clem

. (2005) Addressing solar modulation and long-term uncertainties in scaling secondary cosmic rays for in situ cosmogenic nuclide applications. Earth and Planetary Science Letters 239: 140–161.

71.

McCarroll

(1987) The Schmidt hammer in geomorphology five sources of instrument error. Technical Bulletin: British Geomorphological Research Group 36: 16–27.

72.

McCarroll

(1989a) Potential and limitations of the Schmidt hammer for relative-age dating field tests on Neoglacial moraines, Jotunheimen, southern Norway. Arctic and Alpine Research 21: 268–275.

73.

McCarroll

(1989b) Schmidt hammer relative-age evaluation of a possible pre-‘Little Ice Age’ Neoglacial moraine, Leirbreen, southern Norway. Norsk Geologisk Tidsskrift 69: 125–130.

74.

McCarroll

(1991) The age and origin of Neoglacial moraines in Jotunheimen, southern Norway: New evidence from weathering-based data. Boreas 20: 283–295.

75.

McCarroll

(1994) The Schmidt hammer as a measure of degree of rock surface weathering and terrain age. In: Beck

(ed.) Dating in Exposed and Surface Contexts. Albuquerque, NM: University of New Mexico Press, pp. 29–45.

76.

McCormac

Hogg

Blackwell

. (2004) SHCal04 southern hemisphere calibration 0–11.0 cal kyr BP. Radiocarbon 46: 1087–1092.

77.

McKinzey

Lawson

Kelly

. (2004) A revised Little Ice Age chronology of the Franz Josef Glacier, Westland, New Zealand. Journal of the Royal Society of New Zealand 34: 381–394.

78.

Matthews

Briffa

(2005) The ‘Little Ice Age’: Re-evaluation of an evolving concept. Geografiska Annaler – Series A: Physical Geography 87: 17–36.

79.

Matthews

Dresser

P-Q

(2008) Holocene glacier variation chronology of the Smørstabbtindan massif, Jotunheimen, southern Norway, and the recognition of century- to millennial-scale European Neoglacial events. The Holocene 18: 181–201.

80.

Matthews

McEwen

(2013) High-precision Schmidt-hammer exposure-age dating of flood berms, Vetlestølsdalen, alpine southern Norway: First application and some methodological issues. Geografiska Annaler – Series A: Physical Geography 95: 185–195.

81.

Matthews

Owen

(2010) Schmidt hammer exposure-age dating: Developing linear age-calibration curves using Holocene bedrock surfaces from the Jotunheimen–Jostedalsbreen region of southern Norway. Boreas 39: 105–115.

82.

Matthews

Shakesby

(1984) The status of the ‘Little Ice Age’ in southern Norway: Relative-age dating of Neoglacial moraines with Schmidt hammer and lichenometry. Boreas 13: 333–346.

83.

Matthews

Winkler

(2011) Schmidt-hammer exposure-age dating (SHD): Application to early Holocene moraines and a reappraisal of the reliability of terrestrial cosmogenic-nuclide dating (TCND) at Austanbotnbreen, Jotunheimen, Norway. Boreas 40: 256–270.

84.

Menounos

Osborn

Clague

. (2009) Latest Pleistocene and Holocene glacier fluctuations in western Canada. Quaternary Science Reviews 28: 2049–2074.

85.

Mifsud

Fujioka

Fink

(2012) Extraction and purification of quartz in rocks using hot phosphoric acid for in situ cosmogenic exposure dating. Nuclear Instruments and Methods in Physics Research Section B: Beam Interactions with Materials and Atoms 294: 203–207.

86.

Nesje

(2009) Late Pleistocene and Holocene alpine glacier fluctuation in Scandinavia. Quaternary Science Reviews 28: 2119–2136.

87.

Nesje

Blikra

Anda

(1994) Dating rockfall-avalanche deposits from the degree of rock-surface weathering by Schmidt hammer tests a study from Norangsdalen, Sunnmøre, Norway. Norsk Geologisk Tidsskrift 74: 108–113.

88.

Niedzielski

Migon

Placek

(2009) A minimum sample size required from Schmidt hammer measurements. Earth Surface Processes and Landforms 34: 1713–1725.

89.

Nussbaumer

Steinhilber

Trachsel

. (2011) Alpine climate during the Holocene: A comparison between records of glaciers, lake sediments and solar activity. Journal of Quaternary Science 26: 703–713.

90.

Orombelli

(2011) Holocene mountain glacier fluctuations: A global review. Geografia Fisica e Dinamica Quaternaria 34: 17–24.

91.

Owen

Matthews

Albert

(2007) Rates of Holocene chemical weathering, ‘Little Ice Age’ glacial erosion and implications for Schmidt-hammer dating at a glacier–foreland boundary, Fåbergstølsbreen, southern Norway. The Holocene 17: 829–834.

92.

Porter

(2000) Onset of Neoglaciation in the southern hemisphere. Journal of Quaternary Science 15: 395–408.

93.

Proceq (2004) Operating Instructions (Betonprüfhammer N/NR–L/LR). Zürich: Proceq SA, 17 pp.

94.

Putnam

Denton

Schaefer

. (2010a) Glacier advance in south middle-latitudes during the Antarctic Cold Reversal. Nature Geoscience 3: 700–704.

95.

Putnam

Schaefer

Barrell

DJA

. (2010b) In situ ¹⁰Be production-rate calibration from the Southern Alps, New Zealand. Quaternary Geochronology 5: 392–409.

96.

Putnam

Schaefer

Denton

. (2012) Regional climate control of glaciers in New Zealand and Europe during the pre-industrial Holocene. Nature Geoscience 5: 627–630.

97.

Putnam

Schaefer

Denton

. (2013) The last glacial maximum at 44°S documented by a ¹⁰Be moraine chronology at Lake Ohau, Southern Alps of New Zealand. Quaternary Science Reviews 62: 114–141.

98.

Reuther

(2007) Surface Exposure Dating of Glacial Deposits from the Last Glacial Cycle (Relief Boden Paläoklima 21). Berlin/Stuttgart: Bornträger, 213 pp.

99.

Reznichenko

Davies

Alexander

(2011) Effects of rock avalanches on glacier behaviour and moraine formation. Geomorphology 132: 327–338.

100.

Reznichenko

Davies

Shulmeister

. (2012) Influence of rock avalanches upon the formation of moraines and their subsequent palaeoclimatic interpretation: A critical appraisal. Zeitschrift für Geomorphologie NF, Supplement 56(4): 37–54.

101.

Röthlisberger

(1986) 10.000 Jahre Gletschergeschichte der Erde. Aargau: Sauerlander, 416 pp.

102.

Röthlisberger

Schneebeli

(1979) Genesis of lateral moraine complexes, demonstrated by fossil soils and trunks indicators of postglacial climatic fluctuations. In: Schlüchter

(ed.) Moraines and Varves. Rotterdam: August Aimé Balkema, pp. 387–419.

103.

Sachs

(1999) Angewandte Statistik. 9th Edition. Berlin: Springer, 881 pp.

104.

Schaefer

Denton

Kaplan

. (2009) High-frequency Holocene glacier fluctuations in New Zealand differ from the northern signature. Science 324: 622–625.

105.

Schildgen

Phillips

Purves

(2005) Simulation of snow shielding corrections for cosmogenic nuclide surface exposure dating. Geomorphology 64: 67–85.

106.

Schönwiese

C-D

(1992) Praktische Statistik für Meteorologen und Geowissenschaftler. Berlin/Stuttgart: Bornträger, 231 pp.

107.

Shakesby

Matthews

Owen

(2006) The Schmidt hammer as a relative-age dating tool and its potential for calibrated-age dating in Holocene glaciated environments. Quaternary Science Reviews 25: 2846–2867.

108.

Shakesby

Matthews

Schnabel

(2008) Cosmogenic ¹⁰Be and ²⁶Al ages of Holocene moraines in southern Norway II: Evidence for individualistic responses of high-altitude glaciers to millennial-scale climatic fluctuations. The Holocene 18: 1165–1177.

109.

Shakesby

Matthews

Karlén

. (2011) The Schmidt hammer as a Holocene calibrated-age dating technique: Testing the form of the R-value-age relationship and defining the predicted-age errors. The Holocene 21: 615–628.

110.

Shulmeister

Davies

Evans

DJA

. (2009) Catastrophic landslides, glacier behaviour and moraine formation – A view from an active plate margin. Quaternary Science Reviews 28: 1085–1096.

111.

Solomina

Haeberli

Kull

. (2008) Historical and Holocene glacier-climate variations general concept and overview. Global and Planetary Change 60: 1–9.

112.

Stahl

Winkler

Quigley

. (2013) Schmidt hammer exposure-age dating (SHD) of late Quaternary fluvial terraces in New Zealand. Earth Surface Processes and Landforms. Epub ahead of print 5 May. DOI: 10.1002/esp.3427.

113.

Stone

(2000) Air pressure and cosmogenic isotope production. Journal of Geophysical Research 105(B10): 23753–23759.

114.

Stroeven

Fabel

Harbor

. (2011) Importance of sampling across an assemblage of glacial landforms for interpreting cosmogenic ages of deglaciation. Quaternary Research 76: 148–156.

115.

Vacco

Alley

Pollard

(2010) Glacier advance and stagnation caused by rock avalanches. Earth and Planetary Science Letters 294: 123–130.

116.

Viles

Goudie

Grab

. (2011) The use of the Schmidt hammer and Equotip for rock hardness assessment in geomorphology and heritage science: A comparative analysis. Earth Surface Processes and Landforms 36: 320–333.

117.

Wanner

Beer

Bütikofer

. (2008) Mid- to late Holocene climate change: An overview. Quaternary Science Reviews 27: 1791–1828.

118.

Wanner

Solomina

Grosjean

. (2011) Structure and origin of Holocene cold events. Quaternary Science Reviews 30: 3109–3123.

119.

Wardle

(1973) Variations of the glaciers of Westland National Park and the Hooker Range, New Zealand. New Zealand Journal of Botany 11: 349–388.

120.

Wardle

(1978) Further radiocarbon dates from Westland National Park and the Omoeroa River mouth, New Zealand. New Zealand Journal of Botany 16: 147–152.

121.

Watchman

Twidale

(2002) Relative and ‘absolute’ dating of land surfaces. Earth-Science Reviews 58: 1–49.

122.

WGMS (2008) Global Glacier Changes: Facts and Figures. Zürich: UNEP/World Glacier Monitoring Service, 88 pp.

123.

Winkler

(2000) The ‘Little Ice Age’ maximum in the Southern Alps, New Zealand – Preliminary results at Mueller Glacier. The Holocene 10: 643–647.

124.

Winkler

(2004) Lichenometric dating of the ‘Little Ice Age’ maximum in Mt Cook National Park, Southern Alps, New Zealand. The Holocene 14: 911–920.

125.

Winkler

(2005) The ‘Schmidt hammer’ as a relative-age dating technique potential and limitations of its application on Holocene moraines in Mt Cook National Park, Southern Alps, New Zealand. New Zealand Journal of Geology and Geophysics 48: 105–116.

126.

Winkler

(2009a) First attempt to combine terrestrial cosmogenic nuclide (¹⁰Be) and Schmidt hammer relative-age dating: Strauchon Glacier, Southern Alps, New Zealand. Central European Journal of Geosciences 1: 274–290.

127.

Winkler

(2009b) Gletscher und ihre Landschaften. Darmstadt: Wissenschaftliche Buchgesellschaft Darmstadt, 183 pp.

128.

Winkler

(2012) Addressing geomorphological uncertainties with terrestrial cosmogenic nuclide dating (TCND) – Examples from Holocene moraines in the Southern Alps of New Zealand. Geophysical Research Abstracts 14: EGU2012-1331.

129.

Winkler

Hagedorn

(1999) Lateralmoränen – Morphologie, Genese und Beziehung zu Gletscherstandsschwankungen (Beispiele aus Ostalpen und West-/Zentralnorwegen). Zeitschrift für Geomorphologie NF, Supplement Band 113: 69–84.

130.

Winkler

Matthews

(2010) Holocene glacier chronologies: Are ‘high-resolution’ global and inter-hemispheric chronologies possible? The Holocene 20: 1137–1147.

131.

Winkler

Shakesby

(1995) Anwendung von Lichenometrie und Schmidt-Hammer zur relativen Altersdatierung prä-frührezenter Moränen, am Beispiel der Vorfelder von Guslar-, Mitterkar-, Rofenkar- und Vernagtferner, Ötztaler Alpen, Österreich. Petermanns Geographische Mitteilungen 139: 283–304.

132.

Winkler

Chinn

Gärtner-Roer

. (2010) An introduction to mountain glaciers as climate indicators with spatial and temporal diversity. Erdkunde 64: 97–118.

133.

Woo

Fitzharris

(1992) Reconstruction of mass balance variations for Franz Josef Glacier, New Zealand, 1913 to 1989. Arctic and Alpine Research 24: 281–290.

134.

Zemp

Hoelzle

Haeberli

(2009) Six decades of glacier mass-balance observations: A review of the worldwide monitoring network. Annals of Glaciology 50: 101–111.

Investigation of late-Holocene moraines in the western Southern Alps,New Zealand,applying Schmidt-hammer exposure-age dating

Abstract

Keywords

Introduction

Study area

Methods

Schmidt-hammer exposure-age dating

Terrestrial cosmogenic nuclide (10Be) exposure-age dating

SHD age-calibration curves

Results

TCND and Schmidt-hammer data

SHD age-calibration curves

Discussion

Methodology

Glacier chronology and correlation

Conclusion

Footnotes

Appendix

Acknowledgements

Funding

References

Terrestrial cosmogenic nuclide (¹⁰Be) exposure-age dating