Welsh Journals

THE EFFECT OF WOODLAND STRUCTURE ON BREEDING BIRD COMMUNITIES IN SAMPLE WOODS IN SOUTH-CENTRAL WALES M. E. MASSEY INTRODUCTION By far the most common broad-leaved woodland in Wales is Oak (Quercetum) (Forestry Commission 1952), but approximately 40% of the total in the 1947-49 census was of other types with Ash (Fraxinetum) and Birch (Betuletum) being the most abundant. Comparatively little is known about the breeding bird communities of these Welsh woodlands, the most recently published work being that of Jones (1972), though prior to this Yapp (1962) and Simms (1971) reported the results of transect work. It is widely accepted that the numbers and species of birds breeding in broad-leaved woodland varies from wood to wood, and in Britain it has often been the practice to use the botanical classification of woodlands by dominant tree species composition when describing their breeding bird communities. The work reported in this paper was aimed at discovering whether or not this botanical classification has any obvious significance in relation to breeding bird communities, species range and populations, in an area of south-central Wales. METHODS Sites were chosen as being semi-natural representatives of three of the dominant tree types found in Wales Oak, Ash and Alder (Alnetum) which could be expected to form climax woodland under natural conditions. It was originally planned to include a birchwood in the series, but it was not possible to find a mature birchwood of sufficient size. The other broad- leaved types listed by the Forestry Commission, namely Beech (Fagus sylvatica), Spanish Chestnut (Castanea sativa), Sycamore (Acer pseudoplatanus), and Elm (Ulmus spp), have mostly been introduced to Wales through planting, or their distribution is very restricted as homogeneous woodland and such woods are generally small in area. A fourth type was also chosen and classed as mixed broad-leaved woodland. It is, however, not a very good 'type' example as its heterogeneity probably results from past management, and it is likely that it was originally an oak- wood. It is, nevertheless, representative of many of the smaller woodlands in south-central Wales. Breeding birds were censused by means of the well-known mapping method. For the oakwood, counts were made for the B.T.O. Common Bird Census with twelve morning visits between late April and early June, with those for the other woodlands based upon the same method, but with only six visits. The information obtained was transferred to species maps and the results were analysed to estimate the number of breeding pairs present. This interpretation of field data by the recorder includes that for the oakwood so that the results are more directly comparable with the other woods. For similar reasons a nestbox population of Pied Flycatchers (scientific names are given in the Appendix) and Blue Tits in the oakwood have not been included in the analysis. Instead an estimate, made from knowledge and experience of the local situation, has been made of the expected normal