Glacier snow line variations in the Southern Alps, New Zealand. T. J. Chinn and I. E. Whitehouse

Transcription

1 World Glacier Inventory - Inventaire mondial des Glaciers (Proceedings of the Riederalp Workshop, September 1978; Actes de l'atelier de Riederalp, septembre 1978): IAHS-AISH Publ. no. 126, Glacier snow line variations in the Southern Alps, New Zealand T. J. Chinn and I. E. Whitehouse Abstract. End-of-summer glacial snow line elevation estimates have been made on several hundred glaciers over some 400 km of the Southern Alps from oblique photographs taken over the past two seasons. Aspect correction factors were applied to put all elevations in terms of south-facing glaciers. Accuracy is estimated at ±50 m, but by comparing photographs of successive years very small changes are detectable. The results show that the snow line elevations follow closely the precipitation distribution and exhibit a similar steep west-east gradient, together with a northsouth gradient caused by latitude. The glacial snow line remains slightly below the 'glacial limit snow line' despite extensive glacial recession over the past century. Historical data indicate that over the past half-century glacial snow lines have generally undergone an accelerating increase in elevation which continued until the mid 1970s. Variation de la limite du manteau nival glaciaire dans les Alpes méridionales de la Nouvelle Zélande Résumé. Des estimations de l'altitude de la limite du manteau nival glaciaire de fin d'été ont été faites pour plusieurs centaines de glaciers répartis sur près de 400 km des Alpes méridionales à l'aide de photographies aériennes obliques prises durant les deux saisons écoulées. On s'est servi de facteurs correctifs pour l'exposition, de manière à pouvoir exprimer toutes les altitudes en termes de glaciers exposés au sud. La précision estimée est de ±50 m, mais la comparaison de photographies prises lors de deux années successives permet de détecter de très faibles variations. Les résultats montrent que l'altitude de la limite du manteau nival suit de très près la distribution des précipitations et présente de manière analogue un fort gradient ouest-est, accompagné d'un gradient nord-sud dû à la latitude. La limite du manteau nival glaciaire demeure située légèrement au-dessous de la 'limite du manteau nival de la limite de glaciation' en dépit de la forte récession glaciaire durant le siècle écoulé. Les données historiques indiquent que durant le demi-siècle écoulé l'altitude de la limite du manteau nival glaciaire s'est généralement élevée de manière accélérée jusque vers le milieu des années soixante-dix. INTRODUCTION A New Zealand glacier inventory is being prepared as a secondary aspect of studies of present climate variations. A study of regional trends of glacier snow line elevation forms a part of an alpine climatology project and provides initial data for a glacier inventory. This inventory will enable the areas of permanent snow and ice to be redefined on the New Zealand Land Resource Inventory, where large areas of seasonal snow have been included in this category. All of the presently glaciated portion of the Southern Alps, New Zealand, is being studied. It extends for some 400 km south from the Arthur's Pass area to Fiordland (Figs. 1 and 2). Peaks rise to 3600 m but there is an accordance of summit heights with a dominant peak height of between 2100 and 2300 m. CLIMATE The climate of the whole region is dominated by a succession of anticyclones and troughs of low pressure moving eastwards on to the country from the Tasman Sea. The Southern Alps lie across the path of the prevailing westerly winds, creating a steep eastward precipitation gradient and a strong fôhn effect. Mean annual precipitation rises rapidly from 3000 mm along the narrow west coastal plains to a 219

2 220 T. J. Chinn and I. E. Whitehouse FIGURE glacier snow line elevation contours for the northern section of the glaciated Southern Alps.

3 Glacier snow line variations in New Zealand 221 ' PLOTTING POINTS MAIN DIVIDE FIGURE glacier snow line elevation contours for the southern section of the glaciated Southern Alps.

4 222 T. J. Chinnand I. E. Whitehouse peak of up to mm in the western alps a few kilometres west of the divide (Fig. 4). From this peak, precipitation diminishes approximately exponentially to about 1000 m in the eastern ranges. Although westerly winds prevail, easterly, and more especially, southerly winds are important sources of snow. Precipitation is evenly distributed throughout the year. DATA COLLECTION The data were gathered during a single day on a light aircraft flight from Christchurch. All glaciers seen were photographed using two hand-held medium format (6 X 7) cameras. The timing of the flight, made on 11 April 1978, was planned to coincide approximately with the end of the glacial balance year, but before the first of the winter snowfalls. A very early winter storm during April 1976 forced cancellation of the project for that year. A flight was made the following year on 5 April 1977, but cloudy weather conditions severely reduced the number of glaciers covered. SNOW LINE PLOTTING Identification and plotting the positions of the glaciers photographed constituted a major portion of the project. Over 800 photographs were taken on the 1978 flight, from which 260 snow line elevations were obtained. From the photographs, the position of the flight line was estimated and sketched onto 1: scale maps, and the glaciers identified by comparing the photographs with maps, vertical aerial photographs and use of local knowledge of the region. The outline of each glacier was marked on the map and the position of the glacial snow line sketched in. The 1: maps are contoured at 100 ft (30 m) intervals. Normally the snow line positions could be plotted to within ±30 m. The sum of plotting and map errors give an estimated error of ±50 m for individual gently sloping glacial snow lines. Glacier, snow line and aspect of the equilibrium line area were listed. With nearly half of the glaciers unnamed, referencing glaciers will remain a problem until they are given glacier inventory reference numbers. This study is designed to plot the regional trend of the annual glacier snow line elevations. Complications arise where steep topography introduced large variations in snow line elevations because of avalanche accumulation at lower elevations, and wind and avalanche deflation of snow at higher altitudes. It is therefore necessary to classify the reliability of the data points. The closer the glaciated topography approaches the hypothetically gently sloping surface of the regional snow line, the more reliable the snow line elevation. Plotted elevations were coded for reliability and the positions of the snow line contours were biased towards those points of highest reliability. All snow line positions were critically examined and elevations divided into three categories of reliability based on topography: (1) Gentle gradient, large glaciers (2) Moderate gradient (3) Steep gradient, glacierets Conflicting data within the same small area were scrutinized and the least reliable points ignored. ASPECT Considerable variation in the altitude of the snow line occurs between glaciers of differing aspect. To correct for these variations, the data were transformed to equiva-

5 Glacier snow line variations in New Zealand 223 lent south-facing aspect. To gain correct values, two groups of snow line data (one from the northern Arthur's Pass area, and one from the southern Olivine mountains area) were plotted on a polar diagram (Fig. 3). Mean snow Une elevations were plotted against direction using eight compass points, and one circle for each group constructed through these points (Chinn, 1975). Aspect elevation correlations were then derived. from the mean elevations read off the circle-direction intercepts. Aspect differences are not symmetrical about a N-S axis, the symmetry is displaced to a NNW-SSE axis. This displacement is probably caused by the daily temperature lag. FIGURE 3. Snow line elevation versus aspect polar diagram showing the variation of mean glacial snow line elevation with changes of aspect. The values of elevation corrections derived from Fig. 3 are given in Table 1. Corrections derived from the Arthur's Pass area were employed over the northern section, north from the Whataroa River, while those derived from the Olivine Range were applied to the area to the south of Haast Pass. A mean of the two correction values was applied over the central area (Figs. 1 and 2). The values of these corrections are somewhat higher than those listed by Nogami (1976) in a study of the lower latitude South American Andes.

6 224 T. J. Chinn and I. E. Whitehouse TABLE 1. Snow line variation with aspect Mean difference above south aspect [m] N NE E SE S SW W NW Arthur's Pass Region Central Region Olivines Region RESULTS The glacial snow lines for 1978 for the Southern Alps range in altitude from 1500 m in the south and west to 2200 m in the east. Contours of the 1978 snow line trend surface (Figs. 1 and 2), show a fundamental broad, west-east gradient, extensively deformed by many local variations. Snow lines descend perceptibly from the extreme west to a trough located parallel to, but somewhat east of the zone of maximum precipitation (McSaveney, 1978). With decreasing precipitation, the surface rises rapidly near the crest of the main divide and continues to rise with a lower eastward gradient. Imposed on this generalized pattern are complex topographic variations of two origins. Large mountain masses lying normal to the prevailing northwest winds, and west of the divide, generate precipitation shadow areas of higher snow lines. The most extensive of these crests coincides with the western side of the Landsborough Valley (Fig. 1). This pattern is repeated across the main divide. Where low passes occur, the precipitation shadow effect is drastically reduced and moist maritime air masses can penetrate east of the main divide to depress snow lines. A striking demonstration of this effect is seen by the destruction of the trend surface through the Hollyford Valley where the valley floor elevation is at 485 m. Numerous other lobations occur at other low saddles and passes (Figs. 1 and 2). These lobations follow positions similar to those mapped by Porter (1975a) in a study of the glaciation limit. Ascending gradients of the snow line trend surface meaned normal to the Main Divide vary from 25 m km" 1 to 40 m km -1 (Fig. 5). These gradients are up to four times as steep as those found for the glaciation limit (Porter, 1975a), and ten to twenty times as steep as those measured for arctic and sub-arctic regions (Andrews and Miller, 1972). Localized areas of very steep gradients occur which far exceed the mean values, e.g. 80 m km"" 1 and 125 m km -1 in parts of a section in the Olivine Range area (Fig. 6). The mean northeast to southwest gradient along the Main Divide is 1 m km ~ 1, or Î15 m/degree of latitude. Very limited precipitation data and no temperature data are available for higher altitudes within the alps to compare with the snow line gradients. One study currently in progress is making precipitation measurements at widespread elevations in a transect across the alps in the Rakaia River area (McSaveney, 1978). Figure 4 shows the fundamental west-east pattern of precipitation in a cross section of the mountains, and indicates that the snow line elevations have a complex relationship to this pattern. Although the snow line elevations tend to be inversely proportional to the precipitation, the snow line trough is displaced to the east of the precipitation peak. This displacement presumably results from the interactions of varying amounts of precipitation, cloud cover and temperature. West of the Main Divide, the snow line is lowered by high winter precipitation and frequent cloud cover, while close to the Main Divide, ample summer rain and release of latent heat will have the opposite effect. East of the Main Divide there is less cloud, less winter snow and less summer rain, all factors

7 Glacier snow line variations in New Zealand 225 FIGURE 4. Topographic profiles across the Southern Alps near Whitcombe Pass showing the gradients of both the glacial snow line and annual total precipitation. OISTANCE ( km FIGURE 5. Topographic profile across the Southern Alps north of Mt Cook showing glacial snow line gradients.

8 226 T. J. Chinn and I. E. Whitehouse 30 M> DISTANCE (km) FIGURE 6. Topographical profile across the Southern Alps from the Olivine Range showing glacial snow line gradients. operating to elevate the snow line. Results from two summer studies of components of surface energy balance made on the Ivory Glacier show that net radiative flux dominates and accounts for over 50 per cent of the heat available for ablation (Anderton and Chinn, 1978). The heat content of precipitation accounts for only a small proportion of the ablation, although during storms it was estimated that this source accounted for 45 per cent of the snow pack ablation. Values from the above study are given in Table 2. COMPARISON WITH PAST SNOW LINE Due to incomplete coverage, the trend surface for April 1977 has not been plotted. However, a visual comparison with the photographs for this year shows that in general the 1977 snow lines were slightly lower than in 1978 by an estimated 30 m. This difference would probably be too small to detect in the comparison of two separate annual trend surface maps, but the accuracy obtainable by comparing photographs between successive years makes it possible to prepare accurate maps of differences in annual snow line elevations. From such analyses, requiring only a single day of field work each year, it is possible to compare the gross climate patterns from year to year. TABLE 2. Components of surface energy balance, Ivory Glacier Study period Net radiative flux [%] Sensible heat flux [%] Latent heat flux {%] 5 Jan Feb Jan Feb Heat content of precipitation [%]

9 Glacier snow line variations in New Zealand 227 The glaciation limit mapped by Porter (1975a), normally at m above the glacial equilibrium line positions (0strem, 1966), remains slightly above the glacial snow line despite the extensive rise in snow lines in recent years. Glacial snow line elevations have been recorded during studies on the Tasman and Ivory Glaciers (Chinn and Bellamy, 1970; Anderton and Chinn, 1973; Anderton, 1975, 1976a, 1976b), and are compared with the present snow line in Table 3. During the studies on the Ivory Glacier, the end of season snow line rose above the upper limits of the glacier to an indeterminate elevation. Mass balance figures are given instead. The 1978 value of 1800 m for the Tasman Glacier snow line (unadjusted for aspect), is approximately equal to the mean value of 1806 m for the 13 years of record. TABLE 3. Balance year Glacial snow line variations with time Tasman Glacier elevation [m] Ivory Glacier, area averaged mass balance [m] -2.11* -1.66* Arthur's Pass area [m] * Approximately full year only. ffrom Chinn (1975). Acknowledgements. This paper is published with the permission of the Commissioner of Works. Grateful acknowledgement is made to M. J. McSaveney for a critical review of the manuscript. REFERENCES Anderton, P. W. (1975) Tasman Glacier Annual report no. 33, Ministry of Works and Development, Wellington, New Zealand. Anderton, P. W. (1976a) Ivory Glacier, representative basin for the glacial hydrological region, 197'2-7'3. Annual report no. 35, Ministry of Works and Development, Wellington, New Zealand. Anderton, P. W. (1976b) Ivory Glacier, representative basin for the glacial hydrological region, Annual report no. 36, Ministry of Works and Development, Wellington, New Zealand. Anderton, P. W. and Chinn, T. J. (1973) Ivory Glacier, representative basin for the glacial hydrological region, report no. 2 (April 1969-May 1971) and report no. 3 (May April 1972). Annual report no. 28, Ministry of Worksand Development, Wellington, New Zealand. Anderton, P. W. and Chinn, T. J. (1978) Ivory Glacier, New Zealand, an IHD representative basin study. /. Glacial. 20, no. 82, Andrews, J. T. and Miller, G.H. (1972) Quaternary history of the northern Cumberland Peninsula, Baffin Island, N.W.T., Canada. Part IV: Maps of the present glaciation limits and lowest equilibrium line altitude for north and south Baffin Island. Arct. Alp. Res. 4, Chinn, T.J. (1975) Late Quaternary snowlines and cirque moraines within the Waimakariri watershed. MSc Thesis, University of Canterbury, Canterbury, New Zealand.

10 228 T. J. Chinn and I. E, Whitehouse Chinn, T. J. and Bellamy, R. J. (1970) Ivory Glacier, representative basin for the glacial hydrological region for the period ending 31 December Annual hydrological research report no. 5, Ministry of Works and Development, Wellington, New Zealand. McSaveney, M. J. (1978) The magnitude of erosion across the Southern Alps. Conference on Erosion Assessment and Control, 25 Annual Conference, pp. 1-18: NZ Society of Soil Conservators, Christchurch, New Zealand. Nogami, M. (1976) Altitude of the modern snowline and Pleistocene snowline in the Andes. Geographic report no. 11, Tokyo Metropolitan University, Tokyo. 0strem, G. (1966) The height of the glaciation limit in southern British Columbia and Alberta. Geogr. Ann. 48A, no. 3, Porter, S. C. (1975a) Glaciation limit in New Zealand's Southern Alps. Arct. Alp. Res. 7, no. 1, Porter, S. C. (1975b) Equilibrium line altitudes of Late Quaternary glaciers in the Southern Alps, New Zealand. Quatern. Res. 5, DISCUSSION Meier: We have mapped the altitude of snow Unes following autumn storms using Landsat images (1: ) together with topographic maps. These maps show the same fingers of snow line depression through valleys and other interesting patterns such as those you have obtained. Chinn: We expect that a trend surface through snow line patterns of individual storms would be difficult to apply in New Zealand because the majority of the precipitation falls during 'warm' cyclones with northwesterly wind conditions. Cool southerly winds follow immediately, giving light snowfalls to low altitudes with the snow line elevation partly governed by irregular convective precipitation. 0strem: Have you tried constructing maps of the height of the glaciation limit? I feel that you would get a pattern which is fairly similar to the pattern shown on your illustrations. Inflow of moist air masses will certainly depress the height of the glaciation level. Chinn: The glaciation limit has been mapped for the northern portion of the Southern Alps by Porter and myself. The two maps are very similar but the snow line contours showed more detail in the trend surface. Miiller: What relationship exists between the snow line that you have shown us and the contemporary glacierization of New Zealand? You mentioned mass balance measurements over several years on some selected glaciers. How do your snow lines relate to present and steady state equilibrium lines of glaciers? Chinn: Glacier snow line elevation values since the mid 1960s are available from the Tasman Glacier. The present snow line is at the mean elevation over this period. Miiller: What is the scale of the base maps for the glacier inventory which you are assembling? How many glaciers are there in New Zealand approximately? Chinn: The scale is 1: and there are about 800 glaciers in New Zealand.