Notes concerning the vegetation on deflation surfaces, Kapp Linnk, Spitsbergen JONAS AKERMAN Akerman. J. 1983: Notes concerning the vegetation on deflation surfaces. Kapp LinnC, Spitsbergen. Polar Research I n.s., 161-169. The distribution of vascular plants within four sample areas on a deflation surface on a set of raised beach ridges near Kapp LinnC. Spitsbergen. is described. The \,egetation cover and distribution of the different species are related to the wind pattern and microtopography. Details of the distribution and growth characteristics of Dryas octopetala, Saxifraga oppositifolia, and Silerie acaulis in relation to each other. and to the wind and microtopography are discussed. Jonas i f k e r m a n , Department of Physical Geography, University of Lund. Solvegatan 13, S-22362 Lund. Sweden; January 1983 (revised M a y 1983). The vegetation of Spitsbergen shows a very great regional variability which is mainly dependent upon differences in climate, soils, hydrology, and topography. The vegetation along the strandflats of western Spitsbergen is fairly uniform, however, belonging t o the ‘Dryas zone’ (Hinz 1976). The small-scale vegetation pattern is much more com- plex than the regional, reflecting the extremely variable local conditions which may influence plant growth in the harsh periglacial environment. This paper describes the vegetation pattern on exposed deflation surfaces and its relation to the environmental factors. The study is based o n ‘sec- ondary’ observations made during field work con- cerning geomorphological processes (wind action, patterned ground processes, etc.); it is a brief descriptive report without any deeper analysis or discussion. The investigation area General description T h e investigation area is situated in central west- ern Spitsbergen in the outer part of Isfjorden, east of Kapp Linne and Isfjord Radio Station (78”04’N, 13”38’E) (Fig. 1). To the west, the area is limited by the open sea, t o the north by Isfjor- den, and to the east by the Starostinaksla and Vardeborgaksla mountain ridges. T h e major part of the area is a wide strandflat characterized by a set of raised beach ridges u p to about 6 0 m above sea level. The area is rich in small lakes and ponds and contains several small bogs. As far as is known the area has continuous permafrost and a n active layer varying between 30 cm in the bogs to 2 rn o n the exposed beach ridges. Climate Isfjord Radio Station, which is also a meteoro- logical station, lies in the investigation area. The area has a maritime influenced arctic climate with at least o n e month above *0”C, but below + 10°C. According t o the Koppen classification, the sta- tion has a true tundra climate (ET). The mean annual air temperature is -43°C (1912-1975) and the mean annual precipitation 400.8 m m (1934-1975) (Akerman 1980). T h e annual course of air temperature and precipitation is given in Table 1. The wind climate is of great significance as regards the periglacial processes and the vegeta- tion pattern of the area. In Fig. 2 the basic wind climatic conditions are shown. For further details about the wind climate, refer to Akerman (1980). Site of detail studies The area in which the vegetation studies have been carried o u t is a very level set of beach ridges at a height of 20 to 4 0 m above sea level. It is about 500 X 300 m in extent and is dominated by two very level areas, o n e at a level of 22 to 24 m 162 Jonas Akerniati N ‘r above sea level. and one at between 32 and 38 m above sea level (Figs. 3 and 4). These are deflation surfaces with a few blocks the majority of which are ventifacts showing well developed facets on their northeast exposed sides. The detail mor- phology is dominated by the smooth ridge system and by polygon furrows of two different types (Fig. 5 ) . The large-scale polygon pattern is the result of an ice-Lvedge system with comparatively wide (0.5-1.5 m ) and deep (0.2-0.5 m) polygon furrows. The small-scale polygons are frost fissure POI!- Fig. 1. Orientation map of the investigation area around Kapp L i m e . Place names and heights (in m ) according to N o r s k Polarinstitutt Map No. B9 Blad Isfiorden. scale 1 : 100.00. gons restricted to the active layer, which is here about 2 m d e e p . These furrows are generally nar- rower and shallower than the large-scale pattern. The soils of the area are dominated by a fairly uniform well-washed beach gravel material in the upper 1.5-3111. Below this there are thick gla- ciomarine silty deposits which can be seen from the active thermokarst processes along the shores of the Linnevatnet lake and the Linneelva river. The beach gravel. lacking finer fractions, is well-drained and dry and lacks larger amounts of ground ice i n winter. T u h k I \ I c m rnonrhl! a i r lcmpcrdture ( l Y J 6 1 Y 7 5 ) and mean monthly precipitation (19361975) at Isfjord Radlo Station. Kapp L i n n e . Spitshcrgcn J F M A M J J 4 S 0 N D T e m p ( ” ( ‘ ) - 1 1 7 - 1 1 5 - 1 7 . 2 - 9 . 2 -3.5 + 1 . 6 +4.7 t 4 . 2 t 1 . l -3.4 -7.1 -9.5 Prcc ( r n m ) ? I 5 30 i 30.9 22.3 23.6 24.8 35.8 45.0 40.6 41.3 38.8 35.6 L-/ FEBRUARY I 1 JULY Fig. 2. Monthly percentage frequencies (frequency interval 1 % . black areas represent more than 5 % ) of concurrent wind forces (Beaufort scale) and wind directions at Kapp L i d (Isfjord Radio) during the period 1956 to 1975. Vegetation on deflation surfaces 163 Vegetation characteristics General distribution The study area lacks a continuous vegetation cover and the vegetation and occurrence of iso- lated plants are dictated by the niicrotopography. The regulating factors are the wind and the soil moisture content. T h e general distribution of the vegetation is shown in Fig. 6. In this figure three classes are distinguished: (1) the thermokarst sur- faces, (2) the deflation surfaces, and (3) the shal- low depressions. The tlierrnokarst surfaces are generally very poor in vegetation as they are subject to intensive disturbances like earth flows, slumps and rill erosion. Some areas which are temporarily more stable have mosses. lichens, and some vascular plants - Saxifraga sp., Ranunculus sp., Draba sp., etc. The deflation surfoces constitute the flattest in the area and the crests of the raised beach ridges. These surfaces appear very poor in vegetation, but the number of lichens and mosses is great and the area studied has a fairly large number of wind eroded tufts of Dryas octopetala and Silene acaic- [is. Saxifraga oppositifolia, Draba n i d i s , and Fig. 3. Topographic map of the investigation area. The test sur- faces are shown by squares (A-D). Contour interval 2 r n . 6 0 . 20 10 Stellaria crassipes are also fairly common at pro- tected 'microsites' o n these surfaces. T h e polygon furrows o n the deflation surfaces are also rich in mosses and Salix polaris (Fig. 7 ) . The depressions between the smooth ridges (incl. the polygon furrows) have a protective snow cover during winter. also providing moisture to the soil in the early summer. and have a discon- tinuous vegetation cover dominated by lichens and mosses; the surfaces are comparatively rich in plants. The dominating vascular plants are Salix polaris. Saxifraga oppositifolia, Stellaria crassipes, Silene acaulis, Draba alpina, and Poa alpina. LlNNE ELVA The deflation surfaces T h e deflation surfaces are characterized by the sandy-gravelly material and by a pavement of small stones and pebbles 1-3 cm in diameter (Fig. 8); vegetation coverage is not great. Including the lichens and mosses the percentage of vegetation cover is larger than the general impression may indicate. however. In order to obtain a more representative picture of the vegetation distri- bution, the deflation surfaces were studied in detail within four test surfaces 7 x 4 m large (Fig. 3 ) . These surfaces were photographed from a Fig. 5 . Air photograph of the investigation area. Detail from Sorsk Polarinstitutt airphoto No. S69 2436. Vegetation on deflation surfaces 165 Fig 6 Vegetation character- istics of the investigation area 1 ‘the thermokarst surfaces’, 2 ‘the deflation surfaces’, and 3 ‘the depressions’ - 1 O o U height of 6 m , affording the possibility of obtain- ing the percentage coverage of the vegetation within each 7 x 4 m sample area. Within these surfaces a 1 x l m frame subdivided into an 0.1 X 0.1 m grid (Fig. 7) was used for evaluating the detailed distribution of the vegetation and the number of plants per unit area. T h e most characteristic plant of these surfaces is the Dryas octopetala, which grows in wind eroded tufts evenly distributed over the area. The tufts a r e very characteristically parabolic in shape (Figs, 9 and l l ) , indicating the active wind direc- tion from the northeast (Akerman 1980). T h e percentage coverage of Dryas on the deflation surfaces is 19.4% (Table 2 ) ; the distri- bution arrangement within the sample areas is shown in Fig. 10A-D. The parabolic-shaped Dryas tufts have a very characteristic form which Fig. 7. The characteristic veg- etation pattern in the polygon furrows. The frame is 1 x 1 m. 166 Jonas A k e r m a n has been mapped by stereographic methods based on vertical photographs taken from a height of 2 rn. A n example is given in Fig. 11, Saxifraga opposirifolia, which is one of the most common and characteristic plants of Spitsbergen, is also found o n these deflation surfaces, but here this plant is apparently growing at the very limit of existence. Elsewhere S. opposirifolia is found evenly distributed over almost all suitable sur- faces, but here almost exclusively in association with the Dryas tufts. O n these exposed surfaces S. oppositifolia is obviously entirely dependent upon the protection that the Dryas tufts a r e providing. T h e position of the S . oppositifolia plants within the four test surfaces in association with the Dryus tufts was investigated (Fig. 12 and Table 3). In this investigation six ‘classes’ were distinguished ( A = not in direct contact with the Dryas tufts, B = central lee position, C = marginal lee pos- ition, D = central windward position, E = mar- ginal windward position, a n d F = ‘inside’ the Dryas tufts. T h e distrbution frequence is given in Table 3. It is clearly indicated in this table that S. opposirifolia is dependent upon the Dryas for its existence - 92.3% of all plants o n t h e deflation The Dryas tufts may provide a higher S o i l mois- ture content, a higher content of fines, and above Fig. 8. The parabolic shaped wind-eroded tufts of D q s a oCf0. surfaces are found in ‘Ontact with the Dryas tufts’ perdu on the deflation surfaces. The active uind direction I S indicated by the g u n . Fig. 9. Detail of one perfectly developed parabolic D ~ y a s tuft on the deflation surface. Vegetation on deflation surfaces 167 FLg. 10. The distribution of Dryas octopetala within the four test surfaces based on photo- graphs taken from a height of 6 m above each surface > O C m - Fig. 11. The ’topography’ of one of the parabolic-shaped Drvas tufts on the deflation surface (test surface B) based on ster- eographic measurements in vertical photographs taken from a height of 2 m. Contour interval 2 cm. all protection from the desiccating and abrasive effects of the strong winds, as the high percentage of observations within the ‘lee position class’ (73%) indicates. Another plant which is interesting in relation to wind activity within the area is Silene acaulis. This plant is not very common on the deflation surfaces but occurs in sufficiently large numbers to be used for a simple analysis (Table 4). A A A B E A LD A Fig. 12. The division used in the investigation of the position of Saxifraga oppositifolia in relation to the Dryas octopetala tufts. A . Not in contact with the Dryas tufts, B . central lee position, C. marginal lee position, D . central windward pos- ition, E . marginal windward position, and F. inside the Dryas tufts. Many of the Silene tufts have wind scars on their northeast exposed side (43.6% of the obser- vations) and a clear correlation between the size of the tufts and the wind erosion was observed early. T h e tufts with a diameter below lOcm Table 2. Percentage vegetation cover on the four sample areas based on the total area 7 x 4 m (Dryas) and five 1 x 1 m test squares (Mosses and lichens). A B C D Mean Dryas octopetala 17.9% 20.4‘+ 22.5% 16.7% lY.4% Mosses and lichens 26.6% 19.2% 20,270 28.2% 23.5% Total 44.5% 39 6% 22.7% 44 9% 3 7 . 9 4 168 Jonas Akerman Table 3 The distribution frequence of Saxifraga opposilifofia in relation to tufts of Dryas ocfopctala within four test surfaces on a deflation surface east of Kapp Linnt. Spitsbergen. A B C D Mean A . Not in contact with Dryas 8.4 8 1 7.9 6.3 7.7% B. Central lee position 47.4 44.2 50.3 42.9 46.2% C. Marginal lee position 29.5 28.8 22.7 26.1 26.8% D. Central windward position 4.2 7.2 5.6 9.3 6.6% E . Marginal windward position 2.1 4.5 1.9 3.9 2.7% F. Inside the D r y s tufts 8.4 7 2 17.2 11.5 11 .O% Number of plants ( 7 x 1 m) 95 111 87 105 99.5 Table 4. Observations on the relation between wind erosion and the size of tufts of Silene a c a u l b within the four test surfaces on deflarion surfaces east of Kapp Linne. Spitsbergen. A B C D Mean ~~ Number of tufts (4 x 7 m) Intact Wind eroded Intact < lOcm Intact > lOcm Wind eroded < 10 cm Wind eroded > 10 cm 39 45 22 41 35.7 5 3 . 8 7 ~ 5 1 . 1 9 59.3% 61.2% 56.470 46.2% 48.9% 40.7% 38.870 43.6% 48.7% 444% 49.2% 51,770 48.5% 5 . 2 4 6.7% 10.1% 9.5% 7.9% 7 . 7 4 13.3’3 9.3% 8.6% 9.7% 3 8 . 5 8 35.6cc 3 1 . 4 7 ~ 30.2% 33.970 ‘ T 10. height 9 - 8 . 1 . diameter 0 f i g . 13 The relation between tuft height and tuft diameter of Silene a c a u l b L. at the deflation surfaces, Kapp Linnt. Spitsbergen. ,I 5 10 15 20 25 r25 cm showed erosion in only 7 . 7 % of the observed cases. while tufts with a diameter above lOcm were wind eroded in 38.5% of the observed cases (Table 4). As there is a clear correlation between the tuft diameter and the height of the tufts (Fig. 13), it is indicated that 3-4 cm height is the critical tuft height for Silene acaulis in this environment. Conclusions The distribution of vegetation on the exposed parts of the deflation surfaces shows a character- istic pattern governed by the direct or indirect action of the wind. The microtopography down to the centimetre scale is one of the most impor- tant factors. The dominating vascular plant on the most exposed surfaces is the Dryas octopetala L., which may cover up to 22.5% of the surface. On these surfaces the growth pattern is charac- terized by wind-eroded parabolic-shaped tufts fairly evenly distributed over the surface. Saxifraga oppositifolia L. is the second plant occurring in any quantity and in this environment is apparently entirely dependent upon the pro- tection provided by the Dryas tufts. The majority of the S. oppositifolia L. plants found are located in the lee and with direct contact with these tufts. Silene acaulis, which grows in more or less circular tufts, is commonly subject to wind erosion. Of all observed tufts, 43.6% showed scars of wind erosion. The critical tuft height is about 3-4cm and of tufts higher than 3-4cm, only 7.9% was not eroded by the wind. Vegetation on deflation surfaces 169 References Akerman. H . J. 1980: Studies on periglacial geomorphology in West Spitsbergen. Meddelande f r d n Lunds Uniueristets Geo- grafska Institution Ser. Auh. L X X X I X . 297p. Gill, D. 1972: The point bar environment in the Mackenzie River Delta, Can. J. Earth Sci. 9 ( I l ) , 1382-1393. Hinz, W. 1976: Z u r Okologie der Tundra Zentralspitzbergens. Norsk Polarinstituff Skrifter N r . 163. Peterson. K . M . & Billings, W. D. 1980: Tundra vegetational patterns and succession in relation to microtopography near Atkasook, Alaska. Arctic and Alpine Research 12 ( 4 ) , 473- 482. Raup, H . M. 1971: The vegetation of the Mesters Wig District, Northeast Greenland. Meddelelser om Grlinland. 194 ( 3 ) . Ronning, 0. 1. 1979: Svalbards flora. Norsk Polarinstirutt Polarhbndbok Nr. 1 . Scholander, P. F . 1934: Vascular plants from Northern Sval- bard. Skrifter om Sualbard og Ishaoet N r . 62. Theakstone, W . H . & Knighton, A . D. 1979: The moss Aongs- troemia longipes, an environmentally sensitive colonizer of sediments at Austerdalisen, Norway. Arctic and Alpine Res. 11 ( 3 ) . 353-356.