The Response of New Zealand s Glaciers to Recent Climatic Changes

The Response of New Zealand s Glaciers to Recent Climatic Changes Abstract: The glaciers of the Southern Alps of New Zealand have been studied since the 1800 s. The Little Ice Age (LIA) was a period of glacier maximum, since then there has been a trend of decrease in area and length for all the glaciers that have been studied. This decrease has been driven primarily by climatic changes since this time; however there is uncertainty over whether this is related to temperature, precipitation or average circulation patterns. Individual glaciers have responded differently to this change, some have advanced in recent times due to their unique characteristics. Many factors affect how glaciers respond to climate and there is no one method of assessing the health of an entire region such as the Southern Alps because of this variation. Further research is needed in a number of areas to further our understanding of glaciers in temperate regions. Contents: 1. Introduction 2. Glacier measurement 3. Changes in climate 3.1. Long-term 3.2. Short-term 4. Glacier response to climate 4.1. Long-term 4.2. Short-term 5. Factors affecting glacier response 6. Conclusion 7. References 1. Introduction Glaciers can be used as an indicator of climatic changes, both past and present. As such research has been undertaken into the links between glaciers and climate around the world, however much of this is focused on the Northern Hemisphere. The glaciers of New Zealand have the longest record of observation in the Southern Hemisphere and due to their relatively high sensitivity to climate and temperate maritime location are of global scientific 1

interest (Fitzharris 1999). The last detailed survey carried out in 1978 found that there were 3,144 glaciers of over 1 ha in size covering a total of 1,160 km 2 with a volume of 53km 2 within New Zealand (Chinn 2001). The majority of these are within the Southern Alps of the South Island between 42 o and 46 o S (WGMS 2008) with the flanks of Mt Ruapehu being the only region able to support glaciers in the North Island (Brock et al 2011), see figure 1. The extent of New Zealand s glaciers has fluctuated highly since the beginning of the Holocene approximately 11,500 years ago, with the last maximum extent seen at the end of the Little Ice Age (LIA) (Schaefer et al 2009). This was an intermittent period with average temperatures cooler than that of present occurring between the 16 th and 19 th centuries (Mann 2002). However the date at which individual glaciers reached their maximum during this period varies from 1600 (Chinn et al 2005a) to late 1800 s (Grove 1990) and there is also debate over these dates based on different measurement techniques. For example the maximum of Mueller Glacier has been dated as approximately 1845 by Gellatly et al (1984), whereas Winkler (2002) discredits this and provides an earlier date of 1725 to 1730. Despite this there is consensus that New Zealand s glaciers, along with those in other parts of the world, have decreased in size since the 19 th century (Gellatly 1985, Grove 1990, Chinn 1996, Chinn et al 2005a). A study of 127 glaciers by Chinn et al (1996) found that on average they had shortened by 38% and lost 25% of their area since the end of the LIA, however there was large variability between glaciers in the speed and size of decrease. The main factor behind this decrease is not clear; Salinger et al (1983) write that an increase in temperature is the main course, Hessell et al (1983) a change in precipitation, whereas others propose a more complex interaction between climatic conditions and glacier fluctuation (e.g. Hooker and Fitzharris 1999). The climate of New Zealand is temperate maritime with current average annual temperatures ranging from 15 o C in the far north to 10 o C in the south, and decreasing on average 0.66 o C per 100 m increase in elevation (NZMet 2012). Precipitation is evenly distributed throughout the year, with a country-wide annual average of 600 to 1500 mm (NZMet 2012). In the Southern Alps there is a strong precipitation gradient from west to east because of the orographic affect of the Southern Alps on the prevailing westerlies that bring moist air from the Tasman Sea (Gjermundsen et al 2011). Annual precipitation quoted in the literature for the Southern Alps varies from 2,500 to 4,500 mm on the west coast to between 10,000 and 15,000 mm to the immediate west of the mountain peaks (WGMS 2008, Anderson and Mackintosh 2012, Chinn et al 2005a) this variation is due to difficulties in measuring average precipitation, particularly at higher elevations (Purdie et al 2011a), and the spatial variation between sites used in each study. The climate of New Zealand is strongly influenced by atmospheric circulation patterns such as the El Nino Southern 2

Oscillation (ENSO), the Southern Annular Mode (SAM) and Interdecadal Pacific Oscillation (IPO), which bring about changes in precipitation, temperature, humidity and pressure over different timescales (Brock et al 2011). Not all glaciers respond the same way to a change in climate. Factors such as debris cover, elevation, accumulation area ratio (AAR), steepness, mass balance gradient and presence or absence of proglacial lakes affect an individual glaciers sensitivity and response to a change in climate (Chinn et al 1996). This review aims to use the literature available to determine changes in New Zealand s glaciers since the end of the LIA and the role both long-term and short-term changes in climate have played in this. This will be done by first looking at the way New Zealand s glaciers have been measured and monitored, how the climate has changed both long-term and short-term, how the glaciers have responded to these changes and finally the factors that affect how the glaciers respond to the climate. Figure 1 - Location of glaciers (black circles) and major settlements (black squares) on South Island, New Zealand, inset southwest North Island showing Mt Ruapehu (shaded area) (Source: Chinn et al 1996) 3

2. Glacier measurement Glaciers change as a result of their mass balance. This is the difference between ablation and accumulation and is expressed as a change in length, area and or thickness of the glacier. Therefore measuring these 3 components of a glacier can give an indication of whether the glacier is under positive or negative balance. Because accumulation and ablation are controlled primarily by the climate these changes in a glacier s characteristics can be used to determine how sensitive a particular glacier is to the climate. Mass balance can also be measured directly. The first explorations of New Zealand s glaciers were made in 1862 by von Haast, who made sketch maps, observations and measurements of terminal face height of the glaciers in the Mt Cook region (Grove 1990). Paintings and photographs dating back to the 1870 s, such as those in figure 2, have also been used to piece together the advance and retreat of glaciers in this area (Grove 1990, Gellatly 1985). Using photos and paintings only gives a rough approximation of change. There can be difficultly in comparing images due to different viewpoints and perspectives being used, paintings can be exaggerated for artistic affect and dating unlabelled images can be problematic. Sketch maps if based on accurate measurements can be of more use in determining absolute values for ice thinning and advance or retreat of glacier fronts. Franz Josef glacier has the most complete record of front position measurements, primarily due to its relatively easy accessibility. Glacier front positions give an indication of the balance of a glacier, because a positive balance will cause the glacier to advance and negative to retreat. However there is a delay between a change in mass balance and glacier front movement, related to the glaciers thickness and the ablation at the terminus (Chinn 1996). Terminal moraines are used to determine were a glacier front was in the past and therefore work out its retreat, however as mentioned earlier there can be difficultly dating moraines accurately and not all glaciers produce moraines. Mass balance can be measured directly by making onsite measurements of factors such as solid precipitation to determine accumulation and radiation, heat fluxes and temperature to assess ablation. In order to determine how a glacier has changed in response to the climate long term studies are needed, however due primarily to cost the longest record of mass balance for an entire glacier in New Zealand is only 7 years. This is for Ivory Glacier, on the western side of the Southern Alps, from 1969 to 1975 (Anderton and Chinn 1978). Point specific mass balance s have been calculated for Tasman Glacier, but were not averaged for the whole glacier due to the relatively small number of points compared with the size of the glacier (Anderton 1975; cited in Chinn et al 2005b). Shorter mass balance measurements of 4

2 to 3 years have been carried out on other glaciers (Oerlemans 1997, Anderson and Mackintosh 2006, Hay and Fitzharris 1988). a b c Figure 2 - (a) A painting by Sir William Fox in 1872, (b) a photo taken in c. 1905, both showing sentinel rock to the right and (c) a photo taken in c. 1940, showing the retreat and thinning of Franz Josef Glacier form 1872 to 1940. (Source: Grove 1990) 5

Equilibrium line altitude (ELA) is used as a surrogate for mass balance when onsite measurements are not available (Brock et al 2011). This is the point of the glacier where accumulation and ablation are equal, both at 0. This is inferred from aerial photographs and more recently from high resolution ASTER satellite images using the difference in appearance of fresh snow and firn at the end of season snow line (EOSS) (Chinn et al 2005b, Gjermundsen et al 2011). By measuring the elevation of the ELA and comparing it with the elevation when the glacier is in equilibrium (ELA 0 ) it is possible to tell if the mass balance is positive or negative and how far the glacier is away from equilibrium. Since 1977 annual ELA surveys have been taken of 50 index glaciers in the Southern Alps (Chinn et al 2005b), the annual results are shown in figure 8 and table 1. ELA 0 was initially assumed from the AAR, using an average of 2:1 with the absolute value for each glacier obtained from glacier area-altitude curves. For smaller glaciers the ELA 0 was assumed to be at the glaciers mean elevation. These assumptions affected the accuracy of the results and now ELA 0 is calculated from the mean ELA of each glacier since 1977, therefore the accuracy will improve with each year the study continues. To test the accuracy of the ELA as a proxy for mass balance Chinn et al (2005b) compared the two for Tasman Glacier, however as there was only measurements for fixed sites and the ELA changed annually there was only one year when the two coincided. In 1966/7 the ELA was at 1970m and the point specific mass balance at 1975m was 0. For the over years however they found a correlation of r=0.97 and r=0.81 for sites close to the average ELA. Although this shows that ELA and mass balance correlate well for Tasman Glacier, it may not for other 50 glaciers where mass balance data is not available. Chinn et al (2005b) found that Tasman Glaciers ELA is representative of the Southern Alps as a whole, as shown in figure 3. The EOSS meanders across a glacier and can be intermittent (Chinn 2005b); also snow can be moved across the line by wind and avalanches (Purdie 2011a) this often makes it difficult or sometime impossible to determine the ELA from aerial images. The record for the 50 glaciers is also not complete, there is wide variation in how many of the glaciers are included in each year s average, see table 1, affecting the accuracy of this study for addressing ELA change over all 50 glaciers. However for individual glaciers the ELA can provide an understanding of how it has responded to environmental changes. Purdie et al (2011b) found they could determine ELA from aerial photographs within an accuracy of ±20 m. Modelling has been used in a number of studies (e.g. Anderson and Mackintosh 2006, 2012, Brock et al 2011) to assess how mass balance changes in response to climate. A common problem with these studies and others where direct measurements are not made on the glacier itself is that they use weather data from neighbouring weather stations, often at 6

different elevations to that of the glacier. For example Brock et al (2011) when studying Mangae Huehu Glacier at 2200m to 2500m above sea level on the slopes of Mt Ruapehu used a weather station 10km north at 1097m elevation when the glacier. This is not as accurate as using data from the glacier itself. A constant lapse rate of temperature with elevation is often used in such studies, despite lapse rates varying diurnally, with elevation and depending on the atmospheric conditions (Minder et al 2010, Stone and Carlson 1979, Purdie et al 2011a). This affects the accuracy of the results, however models taking into account a greater number a variables will provide more accurate results. Ice cores have been used to reconstruct climatic conditions and changes in accumulation on glaciers around the world. Purdie et al (2011b) were the first to do it in New Zealand using an ice core from 2004 to reconstruct the past 24 years of accumulation. The use of ice cores on temperate glaciers is more limited than that on polar glaciers because of their relatively fast movement and high turnover. Purdie et al (2011a) point out that uncertainty over compaction rates and amount of snow lost due to wind blow can affect the measurement of accumulation which would also affect ice cores. Figure 3 - Tasman Glacier s ELA compared with that of the average for the Southern Alps, also showing Franz Josef glacier front change since 1958 (Source: Chinn et al 2005b) 7

3. Changes in climate 3.1 Long-term Changes in climatic conditions have an effect on glacier mass balance by affecting ablation and accumulation; however there is uncertainty over whether precipitation, temperature or a combination of both are the driving force behind this (Chin 1996). Most agree that since the end of the LIA the climate of New Zealand has changed however there is debate over the amount of change and the methods being used to measure it. Salinger and Gunn (1975) measured temperature, air pressure and rainfall measurements for 10 locations, 6 urban, 2 rural and 2 small islands, however only include 8 in their report. Not all the sites have the same length of record, for example Auckland s temperature record extends to 1870, whereas Campbell Islands only to 1915. Figure 4 shows the results of 4 of the sites. They found that from 1935 to 1970, despite large annual variation, there was an overall 1 o C increase in temperature when looking at the 20 year running means. Figure 4 - Change in temperature for 4 locations in New Zealand, 2 urban, 1 rural and 1 island showing a general increase from the 1930 s (Source: Salinger and Gunn 1975) They found no significant variation in precipitation. Air pressure was higher in the 1890 s causing weak westerlies, with the migration of the high pressure ridge north in 1894 westerlies became stronger by 1898, westerlies weakened again by 1935. Salinger (1982) also found that there was an increase in average temperatures of 0.5 o C from mid 1940-50 to 1980. 8

The National Institute of Water and Atmospheric Research (NIWA) seven-station temperature series found a 0.91 o C increase in annual average temperature among 7 sites from 1909 to 2009, all of which are in urban areas. As with Salinger and Gunn (1975) they also found large inter-annual variation (NIWA 2010). NIWA was forced to re-evaluate its science due to claims of inaccuracy of its measurements and criticisms of the correction used to account for changes in site location (Gray 2011). After making adjustments they found that the same trend was apparent. Figure 5 shows their results pre and post adjustment. Figure 5 - Previous (blue) and revised (red) average trends in temperature from 1909 to 2009 shown as an anomaly from the 1971 to 2000 average. (Source: NIWA 2010) Hessell (1980) argues that the apparent increases in temperature since 1930 are artificial because the thermometers used have been affected by changes in shelter, screenage and/or urbanisation. He instead suggests that 1890 to 1910 was abnormally cold and that since temperatures have just returned to normal. Gray (2011) argues that the statistical significance of NIWAs findings is very low and that they indicate no significant increase since records began. The urban heat island affect could explain the increase in temperature, Salinger and Gunn (1975) state that there comparisons of Auckland and Christchurch with the rural areas of Waihi and Waimate show that the variations in trends are almost identical, however their results in figure 4 do not appear to match this claim for Christchurch. 9

NIWA also have an eleven-station temperature series which includes urban, rural and island locations which indicates a 1 o C increase from 1931 to 2008 (NIWA 2010), however as with the seven-station series there is missing data. Global average temperatures have increased 0.74 o C from 1905 to 2006 (IPCC 2007), so it could be argued either way that a 0.9 to 1 o C increase in temperature for New Zealand is realistic or not. Chinn (1996) found that the change in ELAs up to the 1970 s corresponded to a 0.6 o C increase in temperature, assuming that there was a constant lapse rate of 0.7 o C per 100m and that precipitation remained unchanged. Chinn later found in 1999 that since the late 1970 s the change in ELAs indicated a cooling of 0.47 o C with the same assumptions, which is counter to NIWAs results. 3.2 Short-term New Zealand s Location in the South Pacific means that its climate is also affected by short term variations in circulation patterns, which in turn can affect glacier mass balances. These circulation patterns and their affect on New Zealand s glaciers are still not fully understood and are the topic of ongoing research (e.g. Ummenhofer et al 2009, Purdie et al 2011b, Brock et al 2011). Westerlies bring moist air from the Tasman Sea, meaning that stronger westerlies usually relate to more precipitation. SAM is a ring of pressure variability between 60 o S and 45 o S which affects the position of the mid-latitude jet stream, which in term affects the strength and frequency of circum-polar westerlies and storm activity over New Zealand (Renwick and Thompson 2006). A positive SAM brings drier warmer conditions to the Southern Alps, around 5mm less per day on average and up to 1.4 o C warmer., whereas a negative SAM means cooler and wetter conditions, particularly west of the Alps. SAM has less affect in the North Island (Brock et al 2011) ENSO, which comprises La Nina, a strengthening of normal Pacific circulation patterns and El Nino, a reversal of these patterns, events also affect the climate of New Zealand. ENSO activity is connected to the Southern Oscillation Index (SOI), a measure of the sea-level pressure differences between Tahiti, French Polynesia, and Darwin, northern Australia (Trenberth 1984). A positive index is linked to La Nina, a weakening of westerly circulation and increase in temperatures, whereas a negative index is linked to El Nino, with the opposite affect (Hooker and Fitzharris 1999, Brook et al 2011). ENSO rotates over a 2 to 8 year period. 10

The IPO also has an effect on the westerlies that reach the South Island of New Zealand. Positive IPO brings stronger westerlies and negative brings weaker (Salinger et al 2001). Other factors such as Tasman Sea surface temperatures and the pressure difference between the North and South Island, often measured between Auckland and Christchurch, also affect weather patterns. 4. Glacier response to climate 4.1 Long-term The most comprehensive study of how glaciers have changed since the end of the LIA was conducted by Chinn (1996, 1999). He found that of the 127 glaciers measured, assumed to have been a representative sample of the 3144 known glaciers, all of them have reduced in length by an average of 38%. Area loss was found to be 25% on average; however this was only based on 25 glaciers because of the difficulties of determining the precise extent during the LIA. Figures 6 and 7 show the results for length and area changes. The glaciers were separated into 3 main types and found variation between these as well as large variations between individual glaciers, the reasons for this will be discussed in section 5. Other studies looking at changes in glacier characteristics focus on a smaller number a glaciers, a particular region or individual glaciers. For example Gellatly et al (1985) studied the glaciers of the Mt Cook region and found that from 1862 to 1982 all of the glaciers had retreated, however like Chinn (1996) there was great variability in the amount of retreat for each glacier, 6 of these glaciers are shown in figure 9. Similarly Gjermunden et al (2011) found that from 1978 to 2002, the glaciers of this region lost an average of 17% of their area, however this was primarily from the large valley glaciers. All three of these studies found that there is a trend of decreasing glacier area and length over time. However for some glaciers this is interspersed with periods of advance. If a glacier has lost length or area it does not necessarily mean that the volume of ice in that glacier has changed, the thickness of the ice also needs to be taken into account. This is hard to determine for the entire length of a glacier unless detailed topographic information is available, because of this many studies do not include it. Thinning has been observed on glaciers including Tasman (Dykes et al 2010) Murchison (Gellatly 1985) and Franz Josef (Grove 1990) since the end of the LIA. 11

Figure 6 - Mean length change for the 127 indexed glaciers of Chinn s (1996) study as a % of LIA extent (black), standard deviation of the mean in white. Figure 7 - Mean area change for 25 of the 127 indexed glaciers of Chinn s (1996) study as a % of LIA area. No area information was available for cirque glaciers. The reason for the overall decrease in length and area is due to primarily negative mass balances since the end of the LIA. There is debate over the cause of this, whether it is an increase in ablation season temperature increasing melt (Salinger and Gunn 1975, Salinger et al 1983, Anderson and Mackintosh 2006, Chinn 1996). Hessel (1983) however found no significant correlation between temperature and mass balance at Franz Josef Glacier, instead he suggested accumulation had changed by way of precipitation. Others believe that glacier fluctuations are due to overall climatic conditions and that the declining trend is due to conditions favouring increased melt and or decreased accumulation becoming more predominant (Brock et al 2011, Purdie et al 2011b, Hooker and Fitzharris 1999). Purdie et al (2011b) suggest that a single hot year can remove multiple years of net accumulation, as Zemp et al (2006) found in the European Alps. This confirms findings by Oerlemans (1992) 12

1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 who found that on Nigardsbreen, a glacier in western Norway that has similar climatic conditions to the western side of the Southern Alps, a 1 o C increase in temperature caused retreat to be twice the advance of a 1 o C decrease in temperature. This suggests an overall increase in average annual temperature since the LIA may not be necessary to cause the widespread retreat of glaciers as inter-annual variability may be more important. Further research is needed. 4.2 Short-term Inter-annual variability in climatic conditions can cause fluctuations in mass balance, which can lead to advance and retreat of a glacier. Chinn found that the average departure from ELA 0 for the glaciers studied since 1977 varied annually and linked these to changes in climatic and environmental conditions; see figure 8 and table 1. Relatively few glaciers have been studied in terms of their individual response to climate, most of which are in the central Southern Alps (Brock et al 2011). 300 200 Mean departure from ELA 0 100 0-100 -200-300 Figure 8 - Mean departure of ELAs from ELA 0 based on the number of glaciers shown in table 1 for each year from 1977 to 2004. Black bars indicate standard deviation. (Based on data from Chinn et al 2005b). Variation has been explained by 1982/3 cooling from El Nino, 1991 cooling from mount Pinnatubo eruption, 1998 warming from switch in IPO (Chinn et al 2005b). Spike in 1990 possibly due to only 2 glaciers being used for the average, so may not be representative of the other 48. 13

Table 1 - Number of glaciers included in each years average, mean change in ELA and standard deviation (Chinn 2005b) Year Number of Readings Mean departure from ELA 0 Standard Deviation 1977 15-19 44 1978 40 38 51 1979 5 30 83 1980 32-22 53 1981 36-42 39 1982 41 7 44 1983 41-131 64 1984 27-109 64 1985 40-77 45 1986 38-16 45 1987 33-37 39 1988 33 12 50 1989 49-4 50 1990 2 239 1991 1-35 1992 15-125 79 1993 49-135 69 1994 50-67 42 1995 50-153 70 1996 48-46 55 1997 45-125 52 1998 50 43 70 1999 48 170 71 2000 46 145 62 2001 50-93 51 2002 49 114 62 2003 49-103 44 2004 45-84 54 Franz Josef Glacier s frontal position shows 2 distinct phases since its LIA maximum. The first of retreat and the second of advance, as shown in figure 9. Hooker and Fitzharris (1999) found that this correlated with a change in climatic conditions. Their findings are summarised in table 2. The increase in temperature was not measured but inferred from changes in the ELA averaged over 20 ablation seasons for both phases. They found no significant difference between the frequency of westerlies, but the difference in strength was significant. They also found that the amount of precipitation was controlled by the strength of the westerlies, increased summer snowfall increased albedo, further lowering melt and stronger westerlies brought cooler temperatures and more cloud. Precipitation was measured at the Franz Josef weather station 7 km north of glacier terminus, corrections were applied to account for a change in site during the study period and partitioning between snow and rain. 14

Figure 9 - Change in Franz Josef terminus from 1894 to 1994. (Source: Hooker and Fitzharris 1999) Table 2 - Climatic differences between the 1954 to 1981/2 retreat phase and 1982 to 1994 advance phase of Franz Josef Glacier (Hooker and Fitzharris 1999) Retreat Phase Warming temperatures during ablation season Less precipitation Southward shift in the subtropical high pressure zone over New Zealand Weaker westerlies during accumulation season More La Nina events Positive sea level pressure anomalies over New Zealand Lower frequency of south-westerlies during ablation season Advance Phase Average 0.2 o C cooler during the ablation season Average 686 mm more in ablation season and 453 mm more in accumulation season A northward shift in the subtropical high pressure zone away from New Zealand Stronger westerlies during accumulation season More El Nino events Negative sea level pressure anomalies over New Zealand Higher frequency of south-westerlies during ablation season Hooker, Grey (Kirkbride and Warren 1997) and Fox glacier also advanced during this period and both Fox and Franz Josef continue to advance, whilst the rest of New Zealand s studied glaciers are in retreat (Gjermudsen et al 2011). 15

Energy for melt (Mj) Purdie et al (2011b) studied the inter-annual variability in accumulation on Tasman Glacier in relation to climate. Like Hooker and Fitzharris (1999) they found that La Nina conditions when SAM was positive led to more negative mass balances, whereas El Nino conditions with negative SAM resulted in more positive balances. However they found when the polarity between SAM and SOI was opposite it lead to lower net accumulation, indicating there is not a linear relationship between these factors and that a possible unknown secondary mechanism may occur at such times. Ummenhofer et al (2009) found that SAM has a stronger influence on New Zealand s climate because net accumulation was high during a period of more frequent La Nina but negative SAM. Brock et al (2011) found that at Mangae Huehu Glacier in the North Island there is no significant correlation between SAM and the surface area of the glacier. Instead ENSO and IPO have a larger influence. Hay and Fitzharris (1988) found that for Ivory glacier the total energy available for melt during the ablation season was dependant on prevailing climatic conditions, altering ablation by up to ±35% of the mean rates. During anticyclones where there is light winds, clear skies and no precipitation the total energy available for melt was lower. During anticyclonic blocking increased precipitation, stronger winds and warmer air provided more energy for melt. Figure 10 shows the high daily variation in energy available for melt and figure 11 shows the average total energy for melt under difference conditions from 2 months in 2 successive ablation seasons. Rain can also increase ablation in the winter, Owens (1984; cited in Hay and Fitzharris 1988) found that a storm delivering 300mm of rain in 12 hours provided up to half the energy available for melt during that period. Figure 10 - Average daily total energy available for melt and components of the energy balance, variation is due to different weather conditions, note the high rainfall (Q R ) on 29 January 1972 increasing the total for that day (Source: Hay and Fitzharris 1988). 16

Energy for melt (Mj) Figure 11 - Average total energy available for melt and the components of the energy balance for different atmospheric conditions (Source: Hay and Fitzharris 1988). These examples show that there are complex interactions between climate, weather and mass balance in New Zealand and that there has not simply been a continued retreat of all glaciers since the LIA. Trenberth et al (1977; cited in Hay and Fitzharris 1988) suggest that the increased north-easterly and decreased south-westerly flows since the 1950 s could explain the increase in ablation and retreat seen in many glaciers since this time (Chinn et al 2005a). 5. Factors affecting glacier response As Chinn (1996) showed not all types of glacier have reduced in length and or area the same amount since the LIA and there is significant variation between glaciers, as indicated by the high value of standard deviation shown in figure 6. There are a range of factors that affect how a glacier responds to the climate. It was initially thought there was a west-east divide in response with those on the west reacting more quickly than those on the east side of the Southern Alps due to differences in weather conditions, primarily the precipitation 17

gradient. Chinn (1996) found that this was not the case and that individual glacier characteristics were more important, this was also found by others including most recently by Anderson and Mackintosh (2012). Purdie et al (2011a) found by comparing accumulation on Franz Josef, west of the divide, and Tasman, east, that although Franz Josef received 29% more snow overall, this only translated to 6% more accumulation, however they admit an ±8% error in their measurements. There was a high level of uncertainty in their results particularly in estimating levels of compaction within snow pits used to measure accumulation and in determining the amount of snow lost through wind blow. They also only measured accumulation at 4 sites, with only 2 of these used in their analysis. Clare et al (2002) also found that the climate varies little between west and east. Chinn (1996) split glaciers into 3 main categories, see figures 6 and 7: Small low shear stress cirque glaciers, these reflect changes in climate almost without delay; Large high shear stress mountain glaciers, which respond to decadal changes in climate and mass balance after several years; Valley glaciers with low gradient tongues that respond more smoothly to secular trends with delays of several decades. These were subdivided into with and without proglacial lakes. Oerlemans (1994) writes that glaciers with a low slope are more responsive to temperature changes than steep glaciers, however in New Zealand Anderson and Mackintosh (2012) state the opposite. The steep valley glaciers Franz Josef and Fox are more responsive than the low angle valley Tasman glacier. They found that debris cover was the main factor, for example the less responsive Burton and Tasman glaciers have around 55% and 30% debris cover, as opposed to Fox and Franz Josef having almost none. Debris cover has been shown to reduce ablation by up to 90%, the amount is dependent on the type of cover. Reznichenko et al (2011) found that the bulk density, porosity, albedo, moisture content and permeability of debris cover all affect how much it protects the ice from solar radiation, therefore insulating it and reducing melt. They found a few cm s of debris will insulate the underlying ice until it warms up sufficiently for the heat to pass through, then ablation will continue as if there was no cover, and may even be greater due to the decreased albedo of cover. However metres thick debris, such as that on the tongue of Tasman Glacier ablation will almost completely cease, however ablation still occurs within the ice which explains why Tasman glacier thinned but not retreated prior to 2000 (Dykes et al 2010). 18

Dyurgerov et al (2009) and Gjermundsen et al (2011) write that AAR is also a major factor controlling the response of glaciers to climate. The accumulation area of Fox and Franz Josef glaciers, for example, is significantly larger than that of neighbouring glaciers in the central Southern Alps, Gjermundsen et al (2011) believe this is part of the reason why these two glaciers have continued to advance, when others in this area have retreated. Allen et al (2008) found that from 2002 to 2006 Stocking and Eugene glaciers advanced, this was put down to their high velocities compared to other glaciers Proglacial lakes also have a strong control other ablation in the Southern Alps. They form when terminal ice levels drop below that of outlet rivers or when supraglacial ponds deepen or become interconnected near the foot of the glacier (Reynolds 2000). Chinn 1996 showed that the relatively slow retreat of glaciers such as Tasman, Maudley, Classen and Godley increased significantly when proglacial lakes formed. Gellatly et al (1985) found that glaciers that had formed proglacial lakes retreated more than those without over the same period, see figure 12. These are all low angle valley glaciers. This is due to processes such as calving that remove large chunks of ice from the foot of a glacier, or by warm water undercutting the ice increasing the melt at the base of the glacier. Supraglacial ponds only form permanently on glaciers with a slope less than 6 o meaning that low angle valley glaciers or glaciers with termini with low slopes are more susceptible to forming proglacial lakes. Dykes et al (2010) reported that Tasman glacier retreated by 34 to 54 m per year from 2000 to 2006. Previously it had retreated very little from its LIA maximum position. While a glacier ends in a proglacial lake it is effectively decoupled from the atmosphere and will continue to retreat until its termini leaves the lake, when it will once again become influenced by the climate. Over factors such as aspect and elevation also play a role in the sensitivity of glaciers to changes in climate (Chinn 1996), however much of the research on the Southern Alps is looking at the area of debris cover and proglacial lakes, which appear to have the largest influence. Despite consensus that the precipitation gradient does not affect sensitivity Anderson and Mackintosh (2012) found that the most sensitive glaciers were those in the region with highest precipitation, including Salisbury and Jalf which have 10m and 9m average precipitation. They admit that detailed topographic information is not available for most glaciers, so this high sensitivity may be due to over factors. But a high precipitation could lead to high mass turnover and therefore enhanced sensitivity to the climate. 19

Figure 12 - Retreat of Mueller, Hooker, and Tasman, without proglacial lakes, and Murchison, Classen and Godley, with proglacial lakes, from 1862 to 1982, all diagrams are drawn to the same scale. (Source: Gelletly et a 1985) 20

6. Conclusion New Zealand s glaciers have a long record of observation and measurement, however it is only relatively recently, since 1977, that constant monitoring of the ELA of 50 index glaciers has occurred. Although there is problems and uncertainties with using ELA as a proxy for mass balance, it provides an indication of the overall behaviour of the glaciers in the Southern Alps. This measurement has been used to piece together a history of the glaciers of New Zealand since the LIA and has shown an overall pattern of retreat. However as there are over 3000 glaciers in the Southern Alps this selective monitoring, although considered to be representative, does not show the whole picture. For example the highly responsive Franz Josef glacier is not included in the 50, despite being one of the most studied glaciers in the Southern Alps. Individual glaciers have responded differently to short-term and longterm changes in climate, some advancing, whilst others retreat. This is down to the complex interactions between glaciers and the climate and is affected by differences such as glacier type, topography, debris cover and AAR, which are unique to each glacier. The overall reduction in glacier area in New Zealand is primarily in response to a changing climate, although there is argument over whether precipitation, temperature or average circulation patterns are the main course of this, there is still relatively little known about IPO, SAM and ENSO s affects on New Zealand. The literature available for this paper is biased towards certain glaciers and a limited number of studies carried out by a few individuals with little conflict in opinions. This is probably due to New Zealand receiving less attention from the scientific community than other areas, which are more easily accessible and have more detailed data on areas such as mass balance and climatic circulation patterns. More research is needed into the characteristics of individual glaciers and their relations to climate. Studies should be carried out on glaciers besides the most studied Franz Josef, Fox and glaciers of Mt Cook National Park and the assumption of representativeness of the 50 index glaciers should be tested further. 21

7. References Allen, S., Owens, I. and Sirguey, P. (2008) Satellite remote sensing procedures for glacial terrain analysis and hazard assessment in the Aoraki Mount Cook region, New Zealand. New Zealand Journal of Geology and Geophysics 51 (1): 73-87 Anderson, B. and Mackintosh, A. (2006) Temperature change is the major driver of lateglacial and Holocene glacier fluctuations in New Zealand. Geology 34: 121-124 Anderson, B. and Mackintosh, A. (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 15pp Anderton, P.W. and Chinn, T.J. (1978) Ivory Glacier, New Zealand; an l.h.d. representative basin study. Journal of Glaciology 20 (82): 67-84 Brock, M.S., Dean, J.F. and Keys, H.J.R. (2011) Response of a mid-latitude cirque glacier to climate over the last two decades: Mangaehuehu Glacier, Mt Ruapehu. Earth Surface Processes and Landforms 36: 1973-1980 Chinn, T., Winkler, S., Salinger, M.J. and Haakensen (2005a) Recent glacier advances in Norway and New Zealand: a comparison of their glaciological and meteorological causes. Geografiska Annaler 87 A (1): 141-157 Chinn, T.J. (1996) New Zealand glacier responses to climate change of the past century, New Zealand Journal of Geology and Geophysics 39 (3): 415-428 Chinn, T.J. (2001) Distribution of the glacier water resources of New Zealand. Journal of Hydrology (NZ) 40 (2): 139-187 Chinn, T.J., Heydenrych, C. and Salinger, M.J. (2005b) Use of the ELA as a practical method of monitoring glacier response to climate in New Zealand's Southern Alps. Journal of Glaciology 51 (172): 85-95 Dykes, R., Brook, M. and Winkler, S. (2010) The contemporary retreat of Tasman Glacier, Southern Alps, New Zealand, and the evolution of Tasman proglacial lake since AD 2000. Erdkunde 64 (2): 141 154 Dyurgerov, M., Meier, M.F. and Bahr, D.B. (2009) A new index of glacier area change: a tool for glacier monitoring. Journal of Glaciology 55 (192): 710-716 22

Fitzharris, B., Lawson, W. and Owens, I. (1999) Research on glaciers and snow in New Zealand. Progress in Physical Geography 23: 469-500 Gellatly, A.F. (1984) The use of rock weathering-rind thickness to re-date moraines in Mount Cook National Park, New Zealand. Arctic and Alpine Research 16: 225-232 Gellatly, A.F. (1985) Historical records of glacier fluctuations in Mount Cook National Park, New Zealand: a century of change. The Geographical Journal 151 (1): 86-99 Gjermundsen, E.F., Mathieu, R., Kääb, A., Chinn, T., Fitzharris, B. and Hagen, J.O. (2011) Assessment of multispectral glacier mapping methods and derivation of glacier area changes, 1978 2002, in the central Southern Alps, New Zealand, from ASTER satellite data, field survey and existing inventory data. Journal of Glaciology 57 (204) 667-683 Gray, V. (2011) The seven station series. Energy & Environment 22 (4): 429-439 Grove, J.M. (1990) The Little Ice Age. New York: Routledge Hay, J.E. and Fitzharris, B.B. (1988)The synoptic climatology of ablation on a New-Zealand glacier. Journal of Climatology 8 (2): 201-215 Hessell, J.W.D. (1983) Climatic effects on the recession of the Franz Josef Glacier. New Zealand Journal of Science 26: 315 320 Hooker, B.L. and Fitzharris, B.B. (1999) The correlation between climatic parameters and the retreat and advance of Franz Josef Glacier, New Zealand. Global and Planetary Change 22: 39-48 IPCC (2007) Summary for policymakers. In: Climate Change 2007: Synthesis Report. Cambridge: Cambridge University Press Kirkbride, M.P. and Warren, C.R. (1997) Calving processes at a grounded ice cliff. Annals of Glaciology 24: 116 121 Mann, M.E. (2002) Little Ice Age. In MacCracken, M.C. and Perry, J.S. eds. Encyclopedia of Global Environmental Change: Volume 1: The Earth system: physical and chemical dimensions of global environmental change. Chichester: John Wiley & Sons, Ltd. Available at: http://www.meteo.psu.edu/~mann/shared/articles/littleiceage.pdf 23

Minder, J.R., Mote, P.W. and Lundquist, J.D. (2010) Surface temperature lapse rates over complex terrain: Lessons from the Cascade Mountains. Journal of Geophysical Research 115: D14122 13pp New Zealand Metrological Service (NZMet) (2012) New Zealand Climate. Available at: http://www.metservice.com/learning/nz-climate [accessed: 24/04/2012] NIWA (2010) NZ temperature record. Available at: http://www.niwa.co.nz/climate/nztemperature-record [accessed: 25/04/2012] Oerlemans, J. (1992) Climate sensitivity of glaciers in southern Norway: application of an energy-balance model to Nigardsbreen, Hellstugbreen and Aifotbreen. Journal of Glaciology 38: 223-232 Oerlemans, J. (1994) Quantifying global warming from the retreat of glaciers. Science 264 (5156): 243-245 Oerlemans, J. (1997) Climate sensitivity of Franz Josef Glacier, New Zealand, as revealed by numerical modelling: Arctic and Alpine Research 29: 233 239 Purdie, H., Mackintosh, A., Lawson, W. and Anderson, B. (2011a) Synoptic influences on snow accumulation on glaciers east and west of a topographic divide: Southern Alps, New Zealand. Arctic, Antarctic, and Alpine Research 43 (1): 82-94 Purdie, H., Mackintosh, A., Lawson, W., Anderson, B., Morgenstern, U., Chinn, T. and Mayewski, P. (2011b) Interannual variability in net accumulation on Tasman Glacier and its relationship with climate. Global and Planetary Change 77 (3-4): 142-152 Renwick, J., Thompson, D. (2006) The Southern Annular Mode and New Zealand climate. Water and Atmosphere 14 (2): 24 25. Reynolds, J.M. (2000) On the formation of supraglacial lakes on debris-covered glaciers. In: Nakawo, M., Raymond, C.F. and Fountain, A. Debris covered glaciers. Wallingford, Oxfordshire: IASH Reznichenko, N.V., Davies, T.R.H. and Alexander, D.J. (2011) Effects of rock avalanches on glacier behaviour and moraine formation. Geomorphology 132: 327-338 Salinger, M.J. (1982) On the suggestion of post-1950 warming over New Zealand. New Zealand Journal of Science 25: 77-86 24

Salinger, M.J. and Gunn, J.M. (1975) Recent climatic warming around New Zealand. Nature 256: 396-398 Salinger, M.J., Heine, M.J. and Burrows, C.J. (1983) Variations of Stocking (Te Wae Wae) Glacier, Mount Cook, and climate relationships. New Zealand Journal of Science 26: 321-338 Salinger, M.J., Renwick, J. and Mullan, A. (2001) Interdecadal Pacific Oscillation and South Pacific climate. International Journal of Climatology 21: 1705 1721 Schaefer, J.M., Denton, G.H., Kaplan, M., Putnam, A., Finkel, R.C., Barrell, D.J.A., Anderson, B.G., Schwartz, R., Mackintosh, A., Chinn, T. and Schlüchter, C. (2009) Highfrequency holocene glacier fluctuations in new zealand differ from the northern signature. Science 324: 622-625 Stone, P.H. and Carlson, J.H. (1979) Atmospheric lapse rate regimes and their parameterization. Journal of the Atmospheric Sciences 36: 415 423 Trenberth, K.E. (1984) Signal versus noise in the Southern Oscillation. Monthly Weather Review 112: 326-332 Ummenhofer, C., Gupta, A. and England, M. (2009) Causes of late twentieth-century trends in New Zealand precipitation. Journal of Climate 22: 3-19 Winkler, S. (2002) The Little Ice Age maximum in the southern Alps, New Zealand: preliminary results at Mueller Glacier. The Holocene 105: 643-647 World Glacial Monitoring Service (WGMS) (2008) Global glacial changes: facts and figures. Zurich: WGMS. Available at: http://www.grid.unep.ch/glaciers/ Zemp, M., Haeberli, W., Hoelzle, M. and Paul, F. (2006) Alpine glaciers to disappear within decades? Geophysical Research Letters 33: L13504 25