5.1 The drifting of both falling and re-suspended snow is a major determinant of the spatial pattern of accumulation. Deposition occurs where the competence of the air to carry its burden of snow falls below a critical threshold, which is directly related to wind speed. Changes in surface roughness, resulting perhaps from an increase in vegetation height, and increases in topographic shelter, both stimulate snow deposition. In the case of the latter, deposition in the form of deep drifts and cornices are most common in sheltered gullies and on leeward slopes (Fohn 1980). In many cases accumulations on leeward slopes in the Scottish mountains can reach depths in excess of 5m which tend to survive as snowbeds into the summer months, long after the more general cover on more exposed slopes has melted. Spink (1980) in his inventory of summer snowbeds over the period 1965-1978 noted, on Lochnagar for example, that large snowbeds survived on north-west facing slopes when snow-bearing winds during the previous winter had been dominantly from east or south-east, although the latest and most persistent snowbeds have remained on slopes of easterly aspect. This was also demonstrated by Watson et al. (1994) in relation to the late snows of the Cairngorm massif. Sant (1994) in a survey of 94 snowbeds in the Scottish Highlands also demonstrated a strong association between wind direction and location on lee slopes. The dominance of westerly air-streams in Scotland was apparent, with all snowbeds located between north-west and south-east, while the altitudinal range and angle of slope emphasise that most frequent locations are in hollows and corries slightly below summit levels.
Figure 5.1: Melting dates of snow beds of Ben Wyvis (a) Coire nan Laogh (805m) and (b) Meall na Speireig (585m) (from Pottie 1995)
5.2 The reasons for the persistence of snowbeds into the late spring and summer months are complex and relate to the antecedent rate of deposition (linked to wind speed and snowfall intensity) and ablation rates (aspect and spring temperatures), processes which vary over very small distances. Data on Scottish snowbeds are not collected routinely. However, data have been collected at specific sites by, for example, Chris Sydes (Sow of Atholl), Adam Watson (Cairngorm), John Pottie (Ben Wyvis) and at various sites by undergraduate students of the University of Stirling. An analysis by Pottie (1995) of the timing of the final disappearance of snowbeds on east-facing slopes on Ben Wyvis revealed no evidence of a long-term change at an altitude of 805m, but at the lower altitude of 585m there was a discernible trend towards earlier dates from the mid-1980's (Figure 5.1). Of particular significance has been the tendency for snowbeds to decrease in frequency and size in recent years, although Watson et al. (1994) suggest that there was little indication of a recognisable long-term trend between 1970 and 1989. However, by 1996 all snowbeds in the UK had melted (Watson et al. 1997), the first summer this had occurred since 1959.
5.3 The association between the species composition of plant communities in the Scottish mountains and the depth and duration of snow cover has long been recognised (Watt and Jones 1948; Poore and McVean 1957). Snow cover affects both the heat energy and moisture balances of the near-surface and sub-surface micro-environment. Although the snow cover reflects a high percentage of photosynthetically active incoming short-wave radiation, some is also transmitted through snow deposit to depths which depends on the crystalline structure. While Curl et al. (1972) determined that up to 5% of radiation penetrated to a depth of 0.15m in an alpine snowfield, O'Neill and Gray (1973) detected reduction to this percentage at the shallower depth of 0.8m in a prairie snow cover. Beneath the snow surface temperature fluctuation is damped, such that the temperature at the underlying surface may be almost constant at 0oC (Oke and Hannell 1966; Gardner 1969), although this depends on the depth and duration of the snow cover (Mackay and Mackay 1974; Goodrich 1982). Oke (1978) suggests that a minimum snow depth of 0.15m is required in order to maintain a relatively consistent ground surface temperature. At depths less than this, the chilling of the snow surface through radiative and sensible heat loss to the atmosphere will penetrate to the underlying ground and vegetation. As vertical temperature gradients through the snow cover are very steep, a relatively small reduction in depth will result in a very marked increase in the risk of frost stress to plant tissues.
5.4 Snowbeds have a particular association with the development of plant communities (Mordaunt 1997). Helm (1982) listed four main factors affecting alpine snow patch vegetation in the Rocky Mountains which were (a) insulation (b) a shortened growing season (c) meltwater and (d) soil movement. To these should be added a reduction in available photosynthetically active solar radiation, the mechanical effects of snow accumulation on steep slopes and the preferential elution of ions from the snowpack. Although a snow cover reduces the amplitude of temperature fluctuations and offers protection against severe frost, desiccation and wind damage, very little insolation is available for photosynthesis. Coker and Coker (1973) found that Phyllodoce caerulea can produce fresh growth under a shallow snow layer, but it is frost damaged very quickly if exposed above the snow surface. Preferential elution of ions from the snowpack has been shown to generate a pulse of highly acid meltwater at the onset of the main melt season which may affect vegetation beneath the snowpack and further downslope (Woolgrove and Woodin 1996)
5.5 Any form of climatic change which involves a general decrease in the amount of snow falling, changes in its persistence and timing, and directional shifts in principal snow-bearing or snow-distributing winds could have a potentially major impact on plant communities at higher elevations in the Scottish Highlands. A future climate in which there is a reduction in the depth and duration of snow cover in the Scottish mountains, while possibly extending the growing season, exposes plants to extremes of temperature and wind abrasion. A more ephemeral winter snow cover could also be detrimental to plant growth, which is stimulated while protected by the cover, but damaged when exposed after snow melt. The snow cover models developed in Chapter Four can provide an estimate of changes in the climatic environment on mountain slopes. The 1961-1990 mean number of days with snow lying indicates that snow cover would be expected on 50 per cent of winter (November-April) days at an altitude of 650m. In contrast, under the Medium-High Scenario for the 2080s, this altitude increases to 900m, which suggests a possible upwards shift of 250m in snow-related environmental zoning along mountain slopes.
5.6 In mountain gullies and corries a change in wind direction during snowfall is likely to result in very different snowpack properties and drift patterns, while a higher frequency of temperature fluctuation around 0oC could lead to a sustained state of saturation in the regolith on slopes below the deepest drifts. Heavier winter precipitation falling as snow in stronger winds in the mountains could lead to the development of deeper, larger and more persistent patches in east-facing hollows (Hill et al. 1999). However, warmer spring temperatures will accelerate melting and it is unlikely that snowbeds at altitudes below 800m will survive over most summer months. The larger and deeper snowbeds at higher altitudes may, however, persist during the occasional summer.
5.7 The apparent sensitivity of montane vegetation to human impacts such as recreation and grazing, and its dependence on the duration of snow-lie, make it likely that changes in climate will have relatively rapid ecological consequences. Difficulties in separating the effects of immediate human impacts and those of regional climatic change are further compounded by uncertainty about the sensitivity of montane ecosystems and the exact nature of the future course of changes in climate at higher altitudes. Changes in vegetation as a result of global climate change are most likely to be seen in a reduction of the area of mildly chionophilous plant communities, expansion of strongly chionophobic communities and a general altitudinal shift upwards of all the communities below about 1000m.