Classification of bog woodland habitat and review and analysis of restoration management at the RSPB Abernethy Forest Reserve

SUMMARY

1. A survey was conducted of 20 discrete bogs in the RSPB Abernethy Forest Reserve. Species data were collected from 300 quadrants, each 2m x 2m and water samples were collected from 200 quadrants. Samples were taken from any relatively homogeneous patches of vegetation identified on the ground without reference to NVC or other vegetation classifications. Published data for a further 142 releves were obtained from the literature.
2. Vegetation maps were prepared for each of the bogs surveyed showing the distribution of trees; evidence of disturbance including drainage, peat cutting, ploughing and planting; and vegetation type.
3. Water samples were analysed by Dr Michael Proctor of the University of Exeter for pH, Na, K, Mg, Ca and optical absorbance at 320 nm.
4. The vegetation data collected as part of this survey were coded using ComKey and classified by TWINSPAN to generate recognisable vegetation types. These were compared with the NVC classification and each type described with respect to floristics, water quality, distribution within Abernethy and the distribution of pine.
5. Data from permanent quadrants recorded in 1992 and 1997 were collated in a database. These quadrants were established to monitor the development of vegetation following restoration work on drained and ploughed land following removal of planted conifers. The data were analysed to identify trends in development of the vegetation and responses to the restoration treatments.
6. Nine discrete vegetation types were identified from the survey data. These represented a range of wet heath, raise and blanket mire and woodland communities which correlate with NVC communities, M15, M18, M19, W4 AND W18. The vegetation types recognised cannot all be assigned with certainty to any one NVC sub-community. This is partly because TWINSPAN will not necessarily draw boundaries between vegetation types that match the NVC boundaries. Some of the vegetation types also probably represent a heterogeneous collection of quadrants of several different types, but with insufficient replication to justify their separation.
7. Pine was widely distributed through all of the vegetation types although mature trees were more abundant on the drier habitats and seedlings were more abundant in the wetter sites. Three terms are proposed to distinguish different populations of pine within these peatland communities:

Woodland Bog – predominantly bog vegetation with abundant pine seedlings due to the heavy seed rain from surrounding woodland. There is high seedling mortality and the few trees that survive are very stunted, heavily diseased and produce rather few seeds.

Wooded Bog – predominantly bog vegetation with scattered mature trees of moderate height but leaving an open canopy. The trees are fertile and often form uneven aged stands through regeneration.

Bog Woodland – predominantly woodland vegetation with tall, dense tree cover on deep peat. The trees are well grown with a dense canopy. A few remnants to bog vegetation remain in the ground flora, though most has been replaced by woodland bryophytes and shrubs.

The bog woodland vegetation was found to be more widespread than previously thought and has probably been underestimated in this survey.

8. Although the TWINSPAN groups do not closely match NVC sub-communities, it was not considered that the presence of pine warrants new community status.