RSPB’s Abernethy Forest Reserve in Strathspey contains the largest single area of native pine in the UK, 1889 ha, amounting to 15.7% of the UK’s native Scot’s pine forest Management is centred on the conservation and extension of the native pinewood ecosystem by allowing the regeneration and development of a self-sustaining forest over the potential woodland area. Densities of red deer (Cervus elaphus) were up to 12 per km2 in 1988/89, too high for such development A fully monitored cull has reduced red deer densities to around 5 per km2. To assess the effects of deer management on forest development, detailed vegetation monitoring was initiated in 1989. By 1992 the percentage of seedlings with evidence of browsing has fallen from 72% to 43% with a concomitant increase of 20% in the total number of established seedlings and saplings over the survey area. Recent regeneration is occurring in all major vegetation types where woodland recovery is most desired, though seedlings remain rare in many open heathland areas.
RSPB Abernethy Forest Reserve
Recognising native pinewood as a priority habitat, the RSPB acquired seven sections of land within Strathspey totalling 12,450 ha between 1975 and 1989 (RSPB, 1991). The RSPB now owns most of the largest remaining tract of intact native Caledonian pinewood (just over 2,900 ha). Of this area, 1030 ha is plantation and 1889 ha native pinewood. The latter constitutes about 50% of the Abernethy SSSI and 15.6% of the remaining 12,000 ha of native pine in Scotland (RSPB, 1994).
As with all other RSPB reserves, a management plan was prepared to formulate policies and objectives, and guide work at the reserve. Three of these policies are:
- To conserve and extend the native pinewood ecosystem for its nationally important assemblages of breeding birds, especially the Scottish crossbill, crested tit, capercaillie, osprey and goldeneye; for its nationally important plant communities, including a number of rare species; and for its nationally important invertebrate populations, including a number of rare species;
- To develop a self-structuring native pine forest over the whole potential woodland area, by leaving existing semi-natural woodland unmanaged; by encouraging natural regeneration of pine and associated broadleaves;
- To allow successful natural regeneration of woodland to continue to develop and expand by maintaining grazing levels within the pinewood and on adjacent moorland at an appropriate level mainly by control of deer numbers. (RSPB, 1994).
Native pinewood is of such limited extent in the UK that natural regeneration in existing areas of Caledonian pine and adjacent heathland should be strongly encouraged (RSPB, 1993). With time this will allow the development of a forest mosaic of pine (Finns sylvestris), silver and downy birch (Betula pendula and B. pubescens), rowan (Sorbus aucuparia), juniper (Juniperus communis), alder (Alnus glutinosa), aspen (Populus tremula), willow (Salix spp.) and, to a lesser extent, holly (Ilex aquifolium). Ultimately, a woodland of a more diverse age structure will develop, with the associated fauna, ground flora and other components representative of typical Caledonian pinewood.
The main factor suppressing woodland development was identified as grazing pressure from both sheep and deer (red Cervus elaphus and roe Capreolus capreolus). Accordingly, seasonal grazing by sheep on the hill ground was stopped in 1990. Deer fences have been removed and therefore deer can move freely within the reserve and between the reserve and neighbouring estates. Fence removal was necessary for a number of reasons, in particular because it can cause high mortality in woodland grouse. Collisions have occurred at a rate of 0.25 and 0.03 per km of fence per month for capercaillie and black grouse, respectively (Cattetal., 1994). With most of the damage perceived to be due to red deer and the difficulty in assessing the population of the more secretive roe deer, most of the following refers to the management of red deer.
The reserve was divided into 8 sections, and co-ordinated counts of deer began in 1988 and 1989. These counts take place in October and March, times when numbers of deer at Abernethy are relatively stable. These first coordinated counts gave densities of deer at 10.7 km2 for October 1988 and 11.7 km2 for March over the entire 8,300 ha count area (see Table 1). Counting the deer on the reserve in this manner gives a ‘snapshot’ result that is an index of the overall numbers of deer using the reserve. Deer numbers have also been assessed by a combination of vantage point counts and deer dung counts using the methodology in Ratcliffe (1987). In brief, transects of 1000 m run along each of the vantage point areas with 7 m x 7 m quadrats every 100 m. Dung pellet groups (a group is a cluster of six or more pellets), are counted for each of the quadrats. These data are then converted into mean densities of deer via a calibration curve utilising the mean dung pellet group count and the expected rate of dung decay. Two initial vantage points were used, point X, an area of 90 ha on the moorland/woodland edge, and point Y, a woodland glade of 125 ha. A third vantage point, Z , was first used in 1994.
Assessment of deer usage of the reserve via counting dung pellet groups may become a more important and practical method in future years as forest development may hinder counts using sighting methods. This method is also advisable for assessing densities of the more secretive and difficult to observe roe deer.
For red deer, cull levels were determined to achieve a rapid reduction in the population. Generally a cull of 15% should keep a red deer population stable (Mitchell et al, 1977).
|Date||Stags||Red Deer Hinds||Calves||Total||Roe Deer Bucks||Does||Kids||Total|
To actually reduce the numbers of red deer found at Abernethy, a cull of 33% of the hinds and calves, and 20% of the stags was set. This high cull of hinds was necessary to ensure a decrease in recruitment in the long term.
Once the deer numbers are compatible with woodland regeneration, the cull will be set at a level that will maintain a stable deer population. The cull is undertaken by fully trained RSPB stalkers, based at Abernethy.
The red deer counts from October 1988 to March 1994 for the Forest Lodge section (8,300 ha) of the reserve are shown in Figure 1. It should be noted that the reduction of red deer numbers by over a third, was not really noticed until after the third winter of culling. The greatest reductions are apparent in the woodland sections and the adjacent heathland with the potential for woodland regeneration.
This figure also shows the change in population structure that has accompanied the reduction in numbers, with hinds now accounting for a lower proportion of the total.
Table 2 summarises the annual deer cull for both red and roe deer from 1988 to 1993.
The results of the dung and vantage point counts for red deer at points X and Y are summarised by Figures 2a and 2b. The confidence limits for each point are given at the 80% level, higher levels would not be advisable for this methodology. Though only a small data set, there is a significant correlation between estimates of red deer density from vantage point and dung counts (P=0.022, r=2.84 df-2 = 8), when both the Y and X data are combined. The dung counts show the largest reduction of deer density to be at the woodland moorland edge, at point X, where density estimates from this method and the vantage point count are consistently below 5 per km2 from 1992 and 1994.
Higher numbers estimated from the vantage point counts in 1994 are due to the deer, mainly stags, remaining in the woodland longer than usual due to the prolonged and severe winter conditions. The stags moved away from the woodland with the onset of fine weather and the low estimates from the dung counts reflect this overall lowering in deer usage.
It is predicted that lower population densities will result in better habitat quality. This in turn may be expected to lead to larger deer, more fecund hinds that produce calves more likely to survive, and stags sporting larger antlers. At Abernethy Forest reserve, all deer carcasses are examined and biometric measurements are taken including sex specific age, weight and fecundity (Corpus luteurri) and/or lactaction in hinds and does. Preliminary analysis of these data (Beaumont, 1993), indicates that an increase in the percentage of two, three and four-year old hinds at the early stages of pregnancy (with Corpus luteum present), from 60%-70% in 1988/89 to 80%-92% in the 1991/92 cull. This may reflect reduced deer density enhancing fecundity, though the data set is from a relatively short period and other environmental conditions can produce such effects.
In 1988 and 1989, four long-term monitoring schemes using permanent quadrats and transects were established across Forest Lodge, the main red deer range within the
reserve, in order to sample all major forest/moorland habitat types.
Detailed baseline data were collected on vegetation communities using the National Vegetation Classification (NVC) scheme (Rodwell, 1991). In the same study, a series of paired, permanent 20 m x 20 m quadrats were established and sampled, (a control plot and a fenced plot) to study the effects of exclusion of grazing animals from woodland communities. The effect of different canopy densities was also taken into account. In all 18 quadrat pairs were established. Full methodology is given in Tickner (1989) and Gilmour (1989). Factors recorded were:
- Vegetation and physical features of each 20 m x 20 m plots as a whole.
- Species composition and percentage cover of plant species in five 2 m x 2 m random quadrats within each 20 m x 20 m plot.
- Enumeration of woody species regeneration.
- Enumeration of tree species present.
- Assignment of height classes to seedlings present.
The survey was repeated in 1992, following the same methodologies (Brereton, 1993). A quicker method of assessing the regeneration over large areas of the reserve was developed and refined (Amphlett, 1993; Hodge, 1993). Basically the area to be surveyed is divided into manageable subunits in which parallel, straight transects were walked, separated by 50m. All seedlings within 2.5m of each transect were recorded. The seedling density per hectare is given by:
5 x transect length in metres
To assess the level of browsing on the regeneration Amphlett undertook a survey of 402 ha of new natural regeneration. This was carried out in April 1994, before the onset of new leader growth. Full details of the methodology are in Amphlett (1994).
In 1989, the surveys revealed a dearth of seedlings and saplings in the areas of existing woodland and also on the adjacent open moorland. Virtually all of the few seedlings encountered were of smaller height than the field layer and this, most probably, was due to browsing. Although still at an early stage, by 1992 the vegetation monitoring results revealed an encouraging response (Table 3). In the woodland monitoring plots in 1989, it was noted that the average frequency of pine seedlings was less than one per 20 m x 20 m plot.
The 1992 repeat survey revealed a 25% increase in the overall number of seedlings, mainly rowan, and an increase in their height. Nearly five times more new seedlings were present in moorland habitat and two and a half times more in undisturbed pinewoods than in artificially disturbed plantations.
Baseline monitoring of heathland transects in 1989 found high levels of browsing and few seedlings. By 1992, the number of seedlings showing evidence of recent browsing had fallen from 72% to 43%. Seedling numbers had increased, though the vast majority were within a few hundred metres of the forest edge, declining in frequency with distance from the forest edge. (Figure 3) Resurvey of the 20 m x 20 m quadrats in 1992 showed a 61% increase in the number of seedlings where grazing is excluded compared with a 17% increase in plots subject to sheep and deer grazing (although sheep were removed in 1990).
The browsing survey revealed that the proportions of samples browsed over the previous winter varied from 5.3% to 40.6%. For the previous 6 to 18 months the proportion browsed was between 12.5% and 63.2%. The reduction of annual height increment following browsing by deer is in the order of 32%. Pine in almost all circumstances achieved some height increment even where browsed. However, the slow rates of growth apparent at Abernethy (a mean annual increment of 7.5 cm per year for unbrowsed pine, n-245), indicate that it may take in excess of 20 years for new seedlings to grow to a height at which the leader is not at risk from browsing. This also assumes continued browsing over this period.
Surveys of habitat and deer populations indicate that at the RSPB's Abernethy Forest Reserve, deer numbers have been successfully reduced in a relatively short time, to levels compatible with establishing regeneration of the native woodland. The accompanying changes in deer and vegetation have been documented and quantified. All management is regularly assessed and revised in the light of new information. Comparisons of all data between 1989 and 1993 have shown the following:
- Dung counts indicate that the red deer population has been reduced by approximately 40% within the main woodland block;
- There has been 20% increase in the total number of established seedlings and saplings over the survey area. This is mainly due to the high recruitment of rowan, which accounts for 68% of all new seedlings;
- The tree seedlings are significantly taller;
- There is evidence of recent tree regeneration in all major vegetation types where woodland recovery is most needed, although seedlings remain rare in many open heathland areas;
- Seedling establishment and growth is most rapid in areas of low canopy cover;
- Seedling growth is relatively slow at Abernethy though almost all pine seedlings exhibited annual growth despite between 12.5% and 63.2% showing signs of being browsed over the preceding 6 to 18 months.
These results illustrate how management, over a relatively short period, has initiated a recovery of the existing pinewood. In the long term, the area of pinewood should increase to its former extent within the boundaries of the reserve and the current woodland should attain a rich and diverse structure more akin to the original wildwood.
The continued gathering of data from the cull will build up a detailed picture of the performance of the deer at Abernethy. In future years trends in morphometric data will be related to changes in population density and habitat quality.
In response to new findings and new initiatives in conservation, additional research on pine woodland regeneration and development will be undertaken at Abernethy. Long term studies into the effects of this regeneration on the aquatic biota within the catchment of the River Nethy are currently underway. These are in collaboration with SOAFD's Freshwater Fisheries Laboratory at Faskally. Also a detailed study of the effects of regeneration on soil properties and the role that soils play in determining where regeneration will take place is being initiated by the Central University of Lancashire.
Such studies, combined with the extent of RSPB research work at Abernethy will go a long way towards aiding greater understanding of our unique native pine woodland ecosystem. It is hoped that RSPB management will stimulate positive conservation management of native pine woodlands on a much wider scale. Only in this way can this internationally important habitat, together with its unique range of wildlife be maintained and extended to some of its former range.
This paper relies heavily on data collected by the staff at RSPB Abernethy Forest Reserve. Much of the initial monitoring, data collection and deer management was instituted with the help of David Lambie (Stalker). Richard Thaxton (Warden of Loch Garten), Andy Amphlett (Cairngorm Tops Warden) and Bob Moncrieff (Assistant Warden, Forest Lodge), have all contributed to the management, monitoring and research. Thanks must also go to the many researchers: lan Bainbridge, Tom Brereton, Ceri Evans, Helen Gilmour, Bob Proctor, Ron Summers and Matthew Tickner to name but a few.
Amphlett, A., 1993: Abernethy Forest Reserve Regeneration Survey, Winter 1992/93. RSPB Internal Report.
Amphlett, A., 1993: A Survey of Browsing of Scots Pine Regeneration. Abernethy Forest Reserve. RSPB Internal Report.
Beaumont, D.J., 1993: Preliminary Analysis of Deer Data. Abernethy Forest RSPB Reserve. RSPB Internal Report.
Brereton, T., 1993: Abernethy Forest Vegetation Research. Tree Regeneration Monitoring Experiments. RSPB Internal Report.
Catt, D.C., Dugan, D., Green, R.E., Moncrieff, R., Moss, R., Picozzi, N., Summers, R.W. and Tyler, G., 1994: Collisions Against Fences by Woodland Grouse in Scotland. Unpubl. Rpt. ITE, Banchory and RSPB, Sandy.
Gilmour, H., 1989: Vegetation Monitoring Schemes at Abernethy to Record Tree Regeneration in Relation to Reduced Deer Grazing Pressure. RSPB Internal Report.
Hodge, J.G.W., 1993: Tree Regeneration at Forest Lodge: A New Method for Estimating Seedling Densities. RSPB Internal Report.
Mitchell, B., Staines, B.W., and Welch, D., 1977: Ecology of Red Deer: A Research Review Relevant to Their Management in Scotland. ITE, Cambridge.
RSPB, 1991: Habitat Action Plan: Caledonian Pine Forest. RSPB, Sandy.
RSPB, 1993: Time for Pine: A Future for Caledonian Pinewoods. RSPB, Sandy.
RSPB, 1994: Management Plan: Abernethy Forest Reserve. RSPB, Sandy.
Ratcliffe, P.R., 1987: The Management of Red Deer in Upland Forests. Forestry Commission Bulletin 71. HMSO, London.
Tickner, M., 1989: Abernethy Forest Reserve: Vegetation Survey. Botanical Site Evaluation and Management Perscriptions. RSPB Internal Report.