1. Introduction
Geochemical records from ice cores provide a realm of information on past climate variability, including information about temperature (Dansgaard 1964), moisture provenance (Araguas-Araguas et al. 2000), atmospheric aerosol loading (Bertler et al. 2005), atmospheric circulation variability (Legrand and Mayewski 1997; Jansson et al. 2007), precipitation (Shiraiwa et al. 2002), sea surface temperature (Jouzel et al. 2005), and anthropogenic pollution (Gabrielle et al. 2008). As such these data give insight into understanding natural climate variability and provide an important baseline against which to assess climate change.
Most research to date has focused on polar ice cores (Jouzel et al. 1987; Legrand and Mayewski 1997; Kreutz et al. 1999), as below freezing temperatures ensure the preservation of primary isotopic/ionic signatures. However, other environments have also been studied, including tropical glaciers (Thompson et al. 2000; Ginot et al. 2002) and glaciers at high-altitude sites, for example, the Altai Mountains (Aizen et al. 2005), Siberia; Yukon Territory, Canada (Holdsworth and Krouse 2002; Fisher et al. 2008); and the Himalaya Mountains, north-central Asia (Kaspari et al. 2008). Even sites at lower elevation with melting, for example, the Canadian Arctic, have been found to contain important climate information (Fisher et al. 1998; Goto-Azuma et al. 2002).
Temperate maritime glaciers have not received as much study, as summer ablation and consequent meltwater percolation (often coupled with large annual snow accumulation), pose significant challenges (Arnason 1981). Yet these sites have the potential to contribute important information, in particular in the Southern Hemisphere, where meteorological records are relatively sparse and short (Jones and Moberg 2003). Already exciting work has been carried out in Patagonia (Shiraiwa et al. 2002), Sweden (Jansson et al. 2007), Norway (Raben and Theakstone 1994; He et al. 2002), and the Canadian Rockies (Sinclair and Marshall 2009). Here we present a systematic evaluation of the potential of New Zealand glaciers for ice core reconstruction.
Orographic processes can modify both isotopic (Guan et al. 2009) and trace element (Gilfedder et al. 2007) signatures, due to preferential rainout and airmass pathways (Dansgaard 1964). Such effects can cause geochemical records of glaciers located on opposite sides of an orographic barrier to show different concentrations and patterns, even if the snow is derived from the same precipitation event. To evaluate this effect in New Zealand ice, we compare data from fresh snow, concurrently collected at two sites during snow precipitation events.
Postdepositional modification of stable water isotopes and trace elements may affect the climate integrity of ice core records, especially on temperate maritime glaciers at relatively low altitude (Raben and Theakstone 1994; Yoshimura et al. 2000; Shiraiwa et al. 2002; Kameda et al. 2003). Previous studies on temperate glaciers have used snowpit sampling at the end of winter to determine if discrete snowstorm event information is maintained (Raben and Theakstone 1994; He et al. 2002). However, changes in snow density, redistribution, and compaction make it difficult to unambiguously identify and correlate snow depth to each precipitation event (Jansson et al. 2007). Here we attempt to reduce this uncertainty by adopting methods that allow clearer identification of snowstorm increments in a snowpit.
Temperate maritime glaciers, like those in New Zealand, experience high annual accumulation (∼6 m w.e., where w.e. means water equivalent) coupled with significant summer melt (∼1 m w.e.) within the accumulation area, posing significant challenges for ice core interpretation (Ruddell and Budd 1990). However, in New Zealand, little analysis has been conducted into the preservation and integrity of isotopic and trace element records stored in glaciers.
We analyzed fresh snow samples, simultaneously collected during snow precipitation, from two temperate glaciers located on opposite sides of New Zealand’s Southern Alps, to determine the climate signal contained in fresh snow and to evaluate how the geochemical signal is modified by the passage of an air mass across the Southern Alps’ orographic barrier. Furthermore, we sample the same precipitation events in the snowpit at the end of the sampling campaign. The two datasets will allow us to do the following:
Contrast isotopic and trace element signatures present in winter snow accumulation on two glaciers located on opposite sides of New Zealand’s Southern Alps.
Establish their potential to provide information on moisture provenance and climate variability.
Quantify the level of signal preservation.
2. Study sites
The New Zealand Southern Alps contain more than 3100 glaciers (Chinn 1996) and have an average altitude of ∼2500 m MSL. The mountain range lies perpendicular to the path of midlatitude westerly winds, resulting in sites having prevailing westerly flow. For this reason, we select two sites from opposite sides of the orographic barrier: the Franz Josef Glacier (FJG) on the western flank of the Southern Alps (43°30′S, 170°14′E) and the Tasman Glacier (TG), immediately on the eastern side of the range (43°30′S, 170°20′E) (Fig. 1). These glaciers have been chosen because they are at similar latitude, and their accumulation areas extend from ∼1900 to 2500 m MSL, enabling study sites to be set up at similar elevation (∼2300 m MSL). Net annual accumulation on these glaciers is around 3–5 m w.e., derived from a combination of large winter snowfall and summer ablation ∼(1–2 m w.e.) at high elevations (Anderson et al. 2006; Purdie et al. 2010a). There are no long-term weather data from these glaciated sites, but annual mean annual temperature at the top of FJG is estimated to be around −4°C (Anderson et al. 2006).
3. Sample collection
During winter 2008 we simultaneously measured snowfall at a snowboard (0.4 m × 0.4 m, with 2-m central pole), in the accumulation area of TG and FJG from 14 July to 6 August. Surface snow samples were collected daily at 0900 h, immediately adjacent to the snowboards, with a clean plastic scoop and deposited directly into sterile 18-oz Whirl-Pak bags. Bags were filled to minimize headspace, secured tightly, labeled, and kept frozen. Additional snow samples were taken after storm cycles and at time intervals within storms when conditions permitted. Snow depth, density, morphology, and temperature were measured daily. In addition, during the same period, automatic weather stations were installed at each site, ∼4 m from the snowboards, to record local temperature, humidity, wind speed, and direction. This sampling strategy allows isotopic and trace element data to be linked directly to the synoptic weather systems from which they derived, and to meteorological data measured on site.
To clearly identify individual snow events being measured at the snowboard, colored wool markers were placed between snowstorm increments. This was done by tying a single thread of wool (∼1 m long) around the central pole and extending the ends of the wool across the snow surface beyond the snowboard boundary. At the end of the study at the FJG site (24 days), a snowpit was dug down to the initial snow surface (the snowboard) and snow samples were collected from between the wool markers. The integrity of the wool markers was checked by calculating the compaction of fresh snow using depth and density measurements, and comparing that to the depth indicated by the markers and the density of snowpit layers. Wind deflation, as measured at the site, was also taken into account (see Purdie et al. 2010b, manuscript submitted to Arct. Antarct. Alp. Res.). The use of the colored wool markers provided a time stamp down the pit and a robust way to identify the snow layers rather than relying solely on density conversion.
Samples were kept frozen during transportation and subsampling took place in the clean laboratory at the GNS Science Ice Core Research Facility. The samples were melted at room temperature and two aliquots were taken, one for stable isotope analysis (δ18O, δD, and deuterium excess), and the other for inductively coupled plasma mass spectrometry (ICP-MS) analysis (major cation, trace metals, and rare earth elements).
4. Analytical techniques
Stable water isotope analysis (δ18O and δD) was conducted at the National Isotope Centre, GNS Science, Lower Hutt, New Zealand. Oxygen stable isotopes were analyzed on a GV Instruments (GVI) AquaPrep attached to a GVI IsoPrime mass spectrometer by the classical equilibration method. A total of 400 μl of water are equilibrated with 3 ml of headspace flushed with CO2 for 24 h at 25.5°C. Carbon dioxide is then extracted and analyzed by dual inlet on the IsoPrime. All oxygen results are reported with respect to Vienna Standard Mean Ocean Water (VSMOW), normalized to internal standards INS11, INS9, and MM1 with reported values of −0.3‰, −17.3‰, and −29.4‰, respectively. All internal standards are routinely measured against VSMOW for quality control and international comparability. The analytical precision of these measurements are ±0.1‰.
Hydrogen stable isotopes were analyzed on a GVI PyrOH attached to a GVI IsoPrime mass spectrometer by direct injection over hot chromium. A total of 5 μL of water was injected into a helium stream through a quartz reactor filled with chromium granules and quartz chips held at 1050°C, where it was reduced to H2 gas. The H2 is then analyzed by continuous flow mode on the IsoPrime. All hydrogen results are reported with respect to VSMOW, normalized to internal standards INS11, INS9, and MM1 with reported values of −3.3‰, −136.5‰, and −231.5‰, respectively. The analytical precision of these measurements are ±1.0‰.
ICP-MS analysis of trace elements was carried out at Victoria University of Wellington on an Agilent 7500cs for 20 trace elements. Samples were acidified prior to analysis to 1% HNO3 using high-purity SEASTAR acid. Samples were transferred to the ICP-MS via an ASX-520 microvolume auto sampler and introduction kit with PFA Teflon nebuliser and quartz spray chamber. Blanks were run after every six samples to establish detection limits (3σ of blanks). Calibration standards were also acidified to 1% HNO3 and prepared daily. The standards were run every 12 samples to correct for machine drift and to calculate overall precision. Elemental concentrations were determined from ICP-MS count data using bracketing analyses of the calibration standards. Reported elemental concentrations comprise the sum of soluble ions and acid-digested particulates. Certified river water reference material (SLRS-4) was run to check data accuracy and ascertain reproducibility.
To identify moisture source regions, back-trajectory analysis was conducted using the Hybrid Single Particle Lagrangian Integrated Trajectory (HYSPLIT) model, available online (at http://ready.arl.noaa.gov/HYSPLIT.php; Draxler and Rolph 2009). National Centers for Environmental Prediction–National Center for Atmospheric Research (NCEP–NCAR) reanalysis data (2.5° grid size) were selected to force the model, and time stamps were set to concur with snowstorm events. To consider synoptic-scale trends in isotopic and trace element data, each day during the study was categorized as 1 of 3 synoptic regimes following the classification of Kidson (2000; Fig. 2). Kidson utilized 40-yr NCEP–NCAR data to identify 12 different weather types for New Zealand, which were further categorized into zonal, troughing, or blocking regimes. Since their development, Kidson indices have been applied to studies of air pollution (Appelhans 2009; Baldi et al. 2009), climate change (Lorrey et al. 2007; B. Mullan 2009, unpublished manuscript), drought patterns (Salinger 2010), and marine biodiversity (Dunn et al. 2009).
5. Results
a. Stable isotopes of water
Stable isotope data are summarized in Table 1. On average δ18O and δD values of fresh snow were more negative at the TG site than at the FJG site by 1.15‰ and 9.24‰, respectively. Overall, the data showed large variability, ranging from approximately −9‰ to −24‰ for δ18O and from approximately −57‰ to −180‰ for δD, resulting in a deuterium excess (d) range of ∼(5–20). This large range suggests that local temperature alone is unlikely to be the dominant influence on the isotopic ratio.
Comparison of pit and fresh snow from FJG showed that in all cases, both δ18O and δD values became less negative over time, with enrichment ranging from 0.26‰ to 2.0‰ for δ18O and from 1.5‰ to 15.3‰ for δD (Fig. 3). In addition, the δ18O–δD slope relation had decreased in the older pit snow, and the amount of change in d averaged 1.03‰.
b. Trace elements
Trace element concentrations [parts per billion (ppb) or parts per trillion (ppt)] in snow samples from both sites were found to be highly variable, with the standard deviation, for most elements, greater than the mean (Table 2). FJG recorded higher average concentrations for 11 of the 20 analyzed elements, including marine species, for example, Na, Mg, Ca, and Sr (Millero 2007). TG snow samples were higher in most terrestrial elements, including Pb, Zr, and V (Gaillardet et al. 2007).
Pearson’s correlation analysis was conducted on trace element data to identify potential relationships. Three elements—Tl, Th, and U—were removed before analysis, as the concentrations of a number of samples fell below detection limits. At FJG marine aerosols (Na, Mg, and Sr) are correlated above the 95% confidence interval, but not so at TG. In contrast, terrestrial elements are highly correlated at both sites: FJG shows high correlation between Ce and La as well as Ca and Rb and TG shows high correlations between La and Ce, as well as Pb and Rb (Table 3).
Principle component (PC) analysis (PCA) of trace element concentrations (using varimax rotation) was also conducted. For FJG two PCs had eigenvalues >1; PC1 accounted for 67% of snow chemistry variance, with the top three loadings on Ti, Rb, and Al. PC2, representing 20% of the variance, was dominated by Mg, Na, and V. At TG four PCs had eigenvalues >1; PC1 explained 40% of variance with top three loadings on terrestrial species Ce, As, and Mn. PC2 explained 18% of variance with loadings on marine elements Mg, Na, and Sr. PC3 explained 12% of variance (Zr, Y, and Mg), and PC4 explained 7% of the variance (Ti, V, and Ba).
6. Discussion
a. Temporal and spatial variability in stable water isotopes and trace elements
In comparing δ18O and δD values between the sites showed in 7 out of 9 occasions when snow was recorded at both sites, TG samples were more negative than FJG samples (Fig. 4). Isotopic depletion in precipitation occurs with increasing altitude, because of the preferential rainout of heavier isotopes, as an air mass passes over an orographic barrier (Dansgaard 1964; Guan et al. 2009). Precipitation also becomes more negative the farther an air mass travels inland, with an increasing deficit in heavier isotopes because of preferential precipitation, often referred to as the continental effect (Dansgaard 1964). Under prevailing westerly flow, an air parcel precipitating at TG has already passed over the Southern Alps, and the air parcel is likely to have precipitated at FJG. Consequently, moisture arriving under these conditions at TG should be isotopically lighter than at FJG, and this was found 78% of the time. Leeside increases in δ18O, because of subcloud evaporation and moisture exchange (Guan et al. 2009), were not expected at TG because of the site’s close proximity to the Southern Alps, and below freezing temperatures. However, on two occasions—23 and 28 July—when snowfall was recorded at both sites, the isotopic trend was reversed. During these days, back-trajectory analysis, 500-mb geopotential height maps, and observed upper-level wind direction indicate that an easterly flow pattern dominated. During such conditions, an air parcel would reach the eastern (TG) site first, thereby reversing orographic effects and preferential rainout. Consequently, snow precipitation during such an event is expected to show depleted values at the FJG site in comparison to the TG site.
While synoptic-scale patterns are strongly modified by large daily variability, caused by local airflow changes, on average, higher δ18O was recorded under zonal flow (Fig. 4). Isotope values were highly variable during blocking and troughing regimes, although a rapid drop in δ18O was recorded during two troughing events on 20 July and 3 August (Fig. 4). Precipitation under zonal flow tended to be lighter, with less aggressive airmass lifting compared to troughing regimes. This may result in slower isotopic rainout compared to faster-moving frontal systems. The sharp drop in δ18O recorded during troughing on two occasions coincided with south-southwest changes in airflow, typical as frontal systems crossed the Southern Alps (Sturman and Tapper 2006).
No statistically significant correlation was found between δ18O and daily average temperature (or snow surface temperature) at either site, despite some days appearing to covary (Fig. 4). Other climate parameters were tested against δ18O, including humidity and wind direction; however, they likewise did not produce any statistically significant relationships. Instead, the data indicate that the isotopic signature on the scale of individual events might be dominated by airmass pathway and moisture source origin.
Despite the strong influences of airmass trajectory, δ18O–temperature relationships have been found on maritime glaciers. Regression analysis between snowpack and storm temperature on glaciers in the Canadian Rocky Mountains resulted in an adjusted correlation coefficient, r 2 = 0.47 (Sinclair and Marshall 2009), and in annual layers on Tasman Glacier, winter-season snow was less enriched than snow from other seasons (Ruddell and Budd 1990). We propose that the lack of relationship found in this study lies in the briefness of our observations. A longer study, encompassing at least a full seasonal cycle, is more likely to detect a temperature–isotopic ratio relationship in New Zealand ice.
Variability in trace element concentration also occurred between sites, but because of the variable nature of data, it was difficult to discern trends between different synoptic conditions (Fig. 4). The western (FJG) glacier site recorded higher concentrations of marine elements (e.g., Na and Sr) with highest concentrations coinciding with zonal flow and a strong northwest storm on 22 July. Prevailing westerly flow explains why FJG tended to have a clearer marine signal, with the signal appearing to dampen as moisture travels across the Southern Alps. Increased travel distance and orographic lifting (Gilfedder et al. 2007) results in the progressive removal of marine elements as the air mass crosses the Southern Alps.
The TG site generally recorded higher concentrations of terrestrial elements (e.g., Pb and V), attributed to the larger landmass over which moisture has to travel to reach the site, and its proximity to industrialized cities. Vanadium, for example, is associated with industry and fossil fuels (Moskalyk and Alfantazi 2003); however, it is also found in schist and greywacke, prominent rock types in the Southern Alps (M. Dow 1998, unpublished manuscript). High correlation between Rb, Ce, and La concentrations at FJG may indicate dust originating from greywacke rock, whereas reduced correlation observed between these elements at TG would suggest a combination of natural and anthropogenic sources.
Lead was often highest east of the Southern Alps, indicating that the Pb source is the more industrialized eastern South Island, although Pb concentrations are high in greywacke rock in comparison to the average crustal abundance (Kennedy and Gadd 2003). High Pb levels were also found in dust samples from ice on the Fox and Franz Josef Glaciers (Marx et al. 2005). After comparison to potential local and Australian sources, Marx et al. (2005) concluded that the Pb was not of New Zealand origin, instead it was derived from long-traveled Australian dust. In our study, high Pb concentrations on the eastern Tasman Glacier appear to coincide with airflow from the southeast or south, indicating that during our observations, some portion of the Pb is likely attributable to a New Zealand source. It is important to note that Pb is a component of aviation fuel, and during winter, the Tasman Glacier accumulation area is subject to regular helicopter landings as part of heli-skiing operations. We find highest Pb concentrations at FJG in the first snow sample, collected soon after helicopter activity. Further research is needed to establish concentrations and dispersion of Pb in relation to local sourced (aircraft activity on these glaciers) and more distant sources from New Zealand and elsewhere.
b. Determining moisture provenance and within storm variability
Tracing moisture source regions and airmass trajectories improves our understanding of regional climate systems and helps to distinguish climate variability from climate trends (Jansson et al. 2007). The d originally defined by Dansgaard (1964) is used as an indicator of precipitation source region and/or changes at the source region (Araguas-Araguas et al. 2000; Schwikowski et al. 2005; Jouzel et al. 2007). Values of d vary seasonally, and generally high values indicate a warm–dry source region and low values a cold–humid source region (Schwikowski et al. 2005). Average d during the study (12.2) was higher than the long-term average (10) recorded for July–August precipitation in Invercargill, situated near the southern tip of the South Island (IAEA 2010).
Large deuterium excess values can be an indication of variability in kinetic fractionation during initial evaporation from the ocean and/or a more complex evaporation history, including reevaporation of moisture from land surfaces and mixing along airmass trajectories (Aizen et al. 2009). High d values recorded from alpine sites have been attributed to effects of moisture recycling and subcloud evaporation (Froehlich et al. 2008). Subcloud evaporation is small during winter, so the high d values recorded at both sites during this study may be attributable to moisture recycling.
Back-trajectory analysis can assist in providing an estimation of moisture source regions (Kahl et al. 1997; Sinclair and Marshall 2009), but absolute certainty is limited by spatial and temporal representivity of meteorological data (Kahl 1993). Even so, compared to upper-level wind direction and back-trajectory analysis (Fig. 5), values of d did appear to be sensitive to moisture provenance. High d values (∼17) were associated with snow likely derived from the warm Tasman Sea (n = 11), and lower values (∼10) were found in snow, which probably originated in the colder–humid Southern Ocean (n = 3). Moisture traveling over the landmass (most likely from a Pacific Ocean source) had intermediate d values (n = 4). It is important to note that sample sizes for the latter two categories are small in comparison to samples from a Tasman Sea source. The predominant snow moisture source during the winter survey appeared to be the Tasman Sea, with 82% (FJG) and 68% (TG) of snowfall measured from this source.
In addition to deuterium excess, variations in trace element composition can also provide a “fingerprint” of moisture source region (Legrand and Mayewski 1997; Gabrielle et al. 2008). Because of the intensive on-site sampling strategy adopted for this study, associations between snow samples and individual storm events were possible.
Moisture source signal was recorded in trace element data from FJG and TG, with a stronger marine signal contained in Tasman Sea storms, and higher levels of pollutants in snow arriving from the Southern or Pacific Oceans (Fig. 5). The source signal was clearer at FJG, which showed a large difference in marine aerosols between the Tasman Sea and other moisture source areas. The smaller marine signal at TG meant there was less difference in those elements between successive storms; however, higher concentrations of terrestrial components were seen in snow from a Southern Ocean source or during easterly flow. Therefore, snow from the eastern side of the Southern Alps could be useful for tracking chances in atmospheric pollutants, while snow from the western side appears to provide a clearer moisture source signal.
c. Postdepositional modification
At FJG enrichment of δ18O and δD was recorded in older pit snow in comparison to its initial fresh value (Fig. 3). This enrichment appeared not to be related to depth (time) or density changes. The 10-cm snow depth temperature (measured daily at 0900 h local time), remained below freezing throughout the study, although positive air temperatures (up to 6.9°C) were recorded on 6 of the 24 days during midafternoon. Three snowpits were dug during the study, and all showed snow temperatures tending toward 0°C at 1.2-m depth. Temperature in the sampled snowpit was −0.8°C at 80-cm depth. Studies in the European Alps (Moser and Stichler 1970) and in Sierra Nevada, California (Taylor et al. 2001), have found that even during winter, metamorphic processes enrich the snowpack. Indeed, such changes can occur rapidly (∼8 days; Moser and Stichler 1970), as snow crystals enlarge and water vapor migrates from convex to concave crystal surfaces (Whillans and Grootes 1985; Dewalle and Rango 2008). Isotopic enrichment can also be driven by evaporation–sublimation (Earman et al. 2006), and even without surface heating, wind ventilation can drive snow sublimation if vapor gradients allow (Neumann and Waddington 2004).
Percolation of meltwater through the snowpack is known to modify both the isotopic and trace element signal, with isotopic fractionation occurring during melt–refreeze processes (Arnason 1969; Hashimoto et al. 2002; Zhou et al. 2008) and elution and mobility of ions by meltwater percolation (Goto-Azuma et al. 1993; Raben and Theakstone 1994; Yoshimura et al. 2000; Shiraiwa et al. 2002; Kameda et al. 2003). It has been suggested that a trend in δ18O can be maintained on temperate glaciers for some months into the summer season, when surface melt refreezes in the snowpack, creating ice layers that prevent further meltwater percolation (He et al. 2002). In fact, preservation of climate information has been demonstrated in cores from the Penny Ice Cap, Baffin Island, Canada, where more than 40% of annual accumulation can be subject to melt–refreeze processes (Fisher et al. 1998; Goto-Azuma et al. 2002).
Results of this study demonstrate high variability in isotope values between successive storms, mixing of old and new snow due to wind redistribution (Purdie et al. 2010b, manuscript submitted to Arct. Antarct. Alp. Res.), plus enrichment from metamorphic processes, which when combined mean that the isotopic signal left in the snowpack at the end of winter can be complex. Over the course of our observations, the shift in isotopic ratios was systematic toward heavier values, which is most likely attributable to a combination of snow metamorphism and wind sublimation. Our extremely high sampling resolution means that even slight offset between actual precipitation of an event and neighboring material that might be mixed during wind scouring can have significant effects. Without sampling over at least one full seasonal cycle, we cannot determine the full effect this can have. However, it is likely that some variability will be smoothed when sampled at seasonal rather than hourly or daily resolution.
7. Conclusions
Differences in stable water isotopes, and trace elements, were recorded in winter snow accumulation on glaciers east and west of the Southern Alps over a 24-day period. The dominant westerly airflow and orographic processes result in snow being isotopically depleted when deposited on the eastern (Tasman) glacier, but this trend is reversed under easterly flow conditions, which are less common. Trace elements were likewise influenced by the dominant westerly flow, with snow on Franz Josef Glacier containing a stronger marine signal and less terrestrial pollutants in comparison to Tasman Glacier. Freshly deposited snow was found to contain information on moisture provenance, in both deuterium excess and trace element variability. We were able to identify three different moisture source groupings (Tasman Sea, Southern Ocean, and overland–Pacific Ocean). From this study, we conclude that a variety of interesting climate information is contained in snow accumulating on New Zealand glaciers and that potential exists for these sites to contribute further to climate research studies.
However, postdepositional modification of signals occurred during the short 24-day period in winter. Such modifications are likely to be more pronounced during summer, including the possibility of substantial melt. While this study provides a high-resolution, systematic, geochemical snapshot, identifying some of the major influences on geochemical records from New Zealand glaciers, clearly more research is required, such as obtaining longer records from either side of the orographic barrier, containing at least one and preferably a number of seasonal cycles.
Acknowledgments
This study was made possible by assistance from a Victoria University doctorate scholarship; Victoria University science faculty, and School of Geography, Environment and Earth Sciences; University of Canterbury Geography Department; New Zealand Mountain Safety Council; New Zealand Alpine Club; Comer Science and Education Foundation; and Foundation for Research, Science and Technology by Victoria University, and GNS Science (Grants VICX0704 and CO5X0202). Thanks to the Department of Conservation, National Institute of Water and Atmospheric Science, Mount Cook Ski Planes, and Fox Glacier Helicopter Services. Lastly, thanks to Jason Watson and Ewan Paterson for their assistance in the field and to Kate Sinclair for her assistance with interpretation.
REFERENCES
Aizen, V. B., E. M. Aizen, K. Fujita, S. A. Nikitin, K. J. Kreutz, and L. N. Takeuchi, 2005: Stable-isotope time series and precipitation origin from firn-core and snow samples, Altai glaciers, Siberia. J. Glaciol., 51 , 637–654.
Aizen, V. B., and Coauthors, 2009: Stable-isotopes and trace element time series from Fedchenko glacier (Pamirs) snow/firn cores. J. Glaciol., 55 , 275–291.
Anderson, B., W. Lawson, I. Owens, and B. Goodsell, 2006: Past and future mass balance of ‘Ka Roimata o Hine Hukatere’ Franz Josef Glacier, New Zealand. J. Glaciol., 52 , 597–607.
Appelhans, T., cited 2009: An air pollution climatology for Christchurch, New Zealand. [Available online at http://www.ide.titech.ac.jp/∼icuc7/extended_abstracts/pdf/374848-1-090514145355-002.pdf].
Araguas-Araguas, L., K. Froehlich, and K. Rozanski, 2000: Deuterium and oxygen-18 isotope composition of precipitation and atmospheric moisture. Hydrol. Processes, 14 , 1341–1355.
Arnason, B., 1969: The exchange of hydrogen isotopes between ice and water in temperate glaciers. Earth Planet. Sci. Lett., 6 , 423–430.
Arnason, B., 1981: Ice and snow hydrology. Stable isotope hydrology: Deuterium and oxygen-18 in the water cycle, J. R. Gat and R. Gonfiantini, Eds., International Atomic Energy Agency (IAEA) Tech. Rep. Series 210, 143–175.
Baldi, M., J. Salinger, K. Dirks, and G. McGregor, 2009: Winter hospital admissions and weather types in the Auckland region. Preprints, Ninth Int. Conf. on Southern Hemisphere Meteorology and Oceanography, Melbourne, VIC, Australia, Bureau of Meteorology, 8 pp. [Available online at http://www.bom.gov.au/events/9icshmo/manuscripts/M1415_Baldi.pdf].
Bertler, N. A. N., P. A. Mayewski, S. B. Sneed, T. R. Naish, U. Morgenstern, and P. J. Barrett, 2005: Solar forcing recorded by aerosol concentrations in coastal Antarctica glacier ice, McMurdo Dry Valleys. Ann. Glaciol., 41 , 52–56.
Chinn, T. J., 1996: New Zealand glacier responses to climate change in the past century. N. Z. J. Geol. Geophys., 39 , 415–428.
Dansgaard, W., 1964: Stable isotopes in precipitation. Tellus, 16 , 436–468.
Dewalle, D., and A. Rango, 2008: Principles of Snow Hydrology. Cambridge University Press, 410 pp.
Draxler, R. R., and G. D. Rolph, cited 2009: HYSPLIT (hybrid single-particle lagrangian integrated trajectory) model. [Available online at http://www.arl.noaa.gov/HYSPLIT.php].
Dunn, M., R. Hurst, J. Renwick, C. Francis, J. Devine, and A. Mckenzie, 2009: Fish abundance and climate trends in New Zealand. New Zealand Aquatic Environment and Biodiversity Rep. 31, 74 pp.
Earman, S., A. Campbell, F. Phillips, and B. Newman, 2006: Isotopic exchange between snow and atmospheric water vapor: Estimation of the snowmelt component of groundwater recharge in the southwestern United States. J. Geophys. Res., 111 , D09302. doi:10.1029/2005JD006470.
Fisher, D., R. Koerner, J. Bourgeois, G. Zielinski, C. Wake, S. Hammer, K. Goto-Azuma, and T. Hondoh, 1998: Penny Ice Cap Cores, Baffin Island, Canada and the Wisconsinan Foxe Dome Connection: Two States of Hudson Bay Ice Cover. Science, 279 , 692–695.
Fisher, D., and Coauthors, 2008: The Mt Logan Holocene–late Wisconsinan isotope record: Tropical Pacific–Yukon connections. Holocene, 18 , 667–677.
Froehlich, K., M. Kralik, W. Papesch, D. Rank, H. Scheifinger, and W. Stichler, 2008: Deuterium excess in precipitation of Alpine regions - moisture recycling. Isot. Environ. Health Stud., 44 , 61–70.
Gabrielle, P., G. Cozzi, S. Torcini, P. Cescon, and C. Barbante, 2008: Trace elements in winter snow of the Dolomites (Italy): A statistical study of natural and anthropogenic contributions. Chemosphere, 72 , 1504–1509.
Gaillardet, J., J. Viers, and B. Dupré, 2007: Trace elements in river waters. Surface and Ground Water, Weathering, and Soils, J. I. Drever, Ed., Vol. 5, Treatise on Geochemistry, Elsevier, 225–272.
Gilfedder, B. S., M. Petri, and H. Biester, 2007: Iodine and bromine speciation in snow and the effect of orographically induced precipitation. Atmos. Chem. Phys., 7 , 2661–2669.
Ginot, P., M. Schwikowski, U. Schotterer, W. Stichler, H. Gaggeler, B. Francou, R. Gallaire, and B. Pouyaud, 2002: Potential for climate variability reconstruction from Andean glaciochemical records. Ann. Glaciol., 35 , 443–450.
Goto-Azuma, K., M. Nakawo, M. Shimizu, N. Azuma, M. Nakayama, and K. Yokoyama, 1993: Temporal changes in chemical stratigraphy of snow cover. Ann. Glaciol., 18 , 85–91.
Goto-Azuma, K., R. Koerner, and D. Fisher, 2002: An ice-core record over the last two centuries from Penny Ice Cap, Baffin Island, Canada. Ann. Glaciol., 35 , 29–35.
Guan, H., C. Simmons, and A. Love, 2009: Orographic controls on rain water isotope distribution in the Mount Lofty Ranges of South Australia. J. Hydrol., 374 , 255–264.
Hashimoto, S., Z. Shiqiao, M. Nakawo, A. Sakai, Y. Ageta, N. Ishikawa, and H. Narita, 2002: Isotope studies of inner snow layers in a temperate region. Hydrol. Processes, 16 , 2209–2220.
He, Y., W. Theakstone, Y. Tandong, C. Guodong, Z. Dian, and Y. Meixue, 2002: Winter-season climatic signals in the recently deposited snowpack on a Norwegian alpine glacier. Nord. Hydrol., 33 , 111–122.
Holdsworth, G., and H. R. Krouse, 2002: Altitudinal variation of the stable isotopes of snow in regions of high relief. J. Glaciol., 48 , 31–41.
IAEA, cited 2010: GIS mapping for analysis, visualization and retrieval of hydrochemical and isotope data for local to global scales [Global Network of Isotopes in Precipitation (GNIP)]. [Available online at http://nds121.iaea.org/wiser/index.php].
Jansson, P., H. Linderholm, R. Pettersson, T. Karlin, and C-M. Mörth, 2007: Assessing the possibility to couple the chemical signal in winter snow on Storglaciären, Sweden, to atmospheric climatology. Ann. Glaciol., 46 , 335–341.
Jones, P., and A. Moberg, 2003: Hemispheric and large-scale surface air temperature variations: An extensive revision and an update to 2001. J. Climate, 16 , 206–223.
Jouzel, J., C. Lorius, J. R. Petit, C. Genthon, N. I. Barkov, V. M. Kotlyakov, and V. M. Petrov, 1987: Vostok ice core: A continuous isotope temperature record over the last climatic cycle (160,000 years). Nature, 329 , 403–408.
Jouzel, J., V. Masson-Delmotte, M. Stievenard, A. Landais, F. Vimeux, S. Johnsen, A. Sveinbjornsdottir, and J. W. C. White, 2005: Rapid deuterium-excess changes in Greenland ice cores: A link between the ocean and atmosphere. C. R. Geosci., 337 , 957–969.
Jouzel, J., and Coauthors, 2007: The GRIP deuterium-excess record. Quat. Sci. Rev., 26 , 1–17.
Kahl, J., 1993: A cautionary note on the use of air trajectories in interpreting atmospheric chemistry measurements. Atmos. Environ., 27A , 3037–3038.
Kahl, J., D. Martinez, H. Kuhns, C. Davidson, J-L. Jaffrezo, and J. Harris, 1997: Air mass trajectories to Summit, Greenland: A 44-year climatology and some episodic events. J. Geophys. Res., 102 , (C12). 26861–26875.
Kameda, T., Y. Fujii, F. Nishio, L. Savatyugin, S. Arkhipov, I. Ponomarev, and N. Mikhailov, 2003: Seasonality of isotopic and chemical species and biomass burning signals remaining in wet snow in the accumulation area of Sofiyskiy Glacier, Russian Altai Mountains. Polar Meteor. Glaciol., 17 , 15–24.
Kaspari, S., R. Hooke, P. Mayewski, S. Kang, S. Hou, and D. Qin, 2008: Snow accumulation rate on Qomolangma (Mount Everest), Himalaya: Synchroneity with sites across the Tibetan Plateau on 50-100 year timescales. J. Glaciol., 54 , 343–352.
Kennedy, P., and J. Gadd, 2003: Preliminary examination of trace elements in tyres, brake pads and road bitumen in New Zealand. Kingett Mitchell Ltd. Rep., 28 pp.
Kidson, J., 2000: An analysis of New Zealand synoptic types and their use in defining weather regimes. Int. J. Climatol., 20 , 299–316.
Kreutz, K. J., and Coauthors, 1999: Seasonal variations of glaciochemical, isotopic and stratigraphic properties in Siple Dome (Antarctica) surface snow. Ann. Glaciol., 29 , 38–44.
Legrand, M., and P. Mayewski, 1997: Glaciochemistry of polar ice cores: A review. Rev. Geophys., 35 , 219–243.
Lorrey, A., A. Fowler, and J. Salinger, 2007: Regional climate regime classification as a qualitative tool for interpreting multi-proxy palaeoclimate data spatial patterns: A New Zealand case study. Palaeogeogr. Palaeoclimatol. Palaeoecol., 253 , 433.
Marx, S., B. Kamber, and H. McGowan, 2005: Provenance of long-travelled dust determined with ultra-trace-element composition: A pilot study with samples from New Zealand glaciers. Earth Surf. Processes Landforms, 30 , 699–716.
Millero, F. J., 2007: Physicochemical controls on seawater. The Oceans and Marine Geochemistry, H. Elderfield, Ed., Vol. 6, Treatise on Geochemistry, Elsevier, 1–21.
Moser, H., and W. Stichler, 1970: Deuterium measurements on snow samples from the Alps. Proc. Symp. on Use of Isotopes in Hydrology, Vienna, Austria, IAEA, 43–57.
Moskalyk, R., and A. M. Alfantazi, 2003: Processing of vanadium: A review. Miner. Eng., 16 , 793–805.
Neumann, T., and E. Waddington, 2004: Effects of firn ventilation on isotopic exchange. J. Glaciol., 50 , 183–194.
Purdie, H., B. Anderson, W. J. Lawson, and A. Mackintosh, 2010a: Spatial variability in snow accumulation on glaciers in the Southern Alps, New Zealand, as revealed by crevasse stratigraphy. Hydrol. Processes, in press.
Raben, P., and W. H. Theakstone, 1994: Isotopic and ionic changes in a snow cover at different altitudes: Observations at Austre Okstindbreen in 1991. Ann. Glaciol., 19 , 85–91.
Ruddell, A. R., and W. F. Budd, 1990: An investigation into the feasibility of using a New Zealand ice core for paleo-climate interpretations. Department of Meteorology, University of Melbourne, 22 pp.
Salinger, J., cited 2010: Drought and extreme temperatures: Future projected climate change. [Available online at http://www.wamis.org/agm/meetings/etdret09/WOS5-Salinger.pdf].
Schwikowski, M., S. Brütsch, A. Palmer, T. M. Jenk, H. W. Gäggeler, and M. Leuenberger, 2005: Identification of precipitation source regions at Fiescherhorn Glacier using the deuterium excess. Annual report 2005. Paul Scherrer Institute Rep., Paper 27. [Available online at http://lch.web.psi.ch/files/anrep05/27.pdf].
Shiraiwa, T., S. Kohshima, R. Uemura, N. Yoshida, S. Motoba, J. Uetake, and M. Godoi, 2002: High net accumulation rates at Campo de Hielo Patagonico Sur, South America, revealed by analysis of a 45.97 m long ice core. Ann. Glaciol., 35 , 84–90.
Sinclair, K. E., and S. J. Marshall, 2009: Temperature and vapour-trajectory controls on the stable-isotope signal in Canadian Rocky Mountain snowpacks. J. Glaciol., 55 , 485–498.
Sturman, A., and N. Tapper, 2006: The Weather and Climate of Australia and New Zealand. 2nd ed. Oxford University Press, 541 pp.
Taylor, S., X. Feng, J. Kirchner, R. Osterhuber, B. Klaue, and C. Renshaw, 2001: Isotopic evolution of a seasonal snowpack and its melt. Water Resour. Res., 37 , 759–769.
Thompson, L., E. Mosley-Thompson, and K. Henderson, 2000: Ice-core palaeoclimate records in tropical South America since the Last Glacial Maximum. J. Quat. Sci., 15 , 377–394.
Whillans, I. M., and P. M. Grootes, 1985: Isotopic diffusion in cold snow and firn. J. Geophys. Res., 90 , 3910–3918.
Yoshimura, Y., S. Kohshima, N. Takeuchi, K. Seko, and K. Fujita, 2000: Himalayan ice-core dating with snow algae. J. Glaciol., 46 , 335–340.
Zhou, S., M. Nakawo, S. Hashimoto, and A. Sakai, 2008: The effect of refreezing on the isotopic composition of melting snowpack. Hydrol. Processes, 22 , 873–882.
Map showing the location of the Southern Alps and the Tasman and Franz Josef Glacier study sites.
Citation: Journal of Climate 23, 18; 10.1175/2010JCLI3701.1
Examples of (a) zonal flow (15 Jul), (b) troughing (22 Jul), and (c) blocking (30 Jul) regimes at 0000 UTC during the study period. Synoptic analysis charts provided by Metservice. Kidson diagrams, showing composite patterns of 1000-hPa height, are from Kidson (2000).
Citation: Journal of Climate 23, 18; 10.1175/2010JCLI3701.1
Changes in isotope concentrations between fresh (solid) and pit (hollow) snow over a ∼20-day period at FJG during winter.
Citation: Journal of Climate 23, 18; 10.1175/2010JCLI3701.1
Daily variability in (left–right and bottom to top) δ18O–snow depth, d– temperature, Mg–Sr, V–Pb at FJG (dark gray) and TG (light gray) during the study period under the different synoptic regimes of Kidson (2000). Moisture source regions for specific storm events are noted above snow accumulation data.
Citation: Journal of Climate 23, 18; 10.1175/2010JCLI3701.1
Back-trajectory airflow analysis (2300 m MSL) showing different water vapor pathways during snowstorm events for (a) Tasman Sea, (b) overland, and (c) Southern Ocean. The d recorded in snow samples from both sites on these days is shown in the bottom left corner of each map.
Citation: Journal of Climate 23, 18; 10.1175/2010JCLI3701.1
Summary of δ18O, δD, and d data for fresh surface snow samples from FJG and TG during winter 2008. (
Trace element data, including detection limits, average concentrations (
Pearson’s correlation coefficients of trace element chemistry for FJG (n = 29) and TG (n = 45) snow samples. Correlations significant at 0.05 (two tailed) are in italics. Correlations significant at 0.01 (two tailed) are bold.
