Ecological Association of Bird Species and Habitats in Eastern Australia; Similarity Analysis

Abstract

Ornithologists have long recognized the fact that bird species tend to be associated with particular habitats. The terrestrial habitats are generally described in terms of vegetation or land use (Palmgren 1930; Moreau 1934; Lack & Venables 1939; Kendeigh 1944), and the species-habitat associations are analysed in terms of ecological and behavioural habitat selection of birds (Lack 1933; Svardson 1949; Klopfer & Hailman 1965). The basis for classification of bird habitats is usually provided by the structural forms of vegetation, which are reflected by the ecological distribution of bird species, However, patterns of resource utilization by birds are complex and are never completely matched by hierarchical arrangements of plant communities. Thus, recognized assemblages of bird species for a given habitat may not share the boundaries with plant communities within major formation types (Kendeigh 1948, 1954), nor is it possible to predict precisely the distribution and abundance of species in relation to any particular features of vegetation (MacArthur 1964). Bond (1957) studied the ecological distribution of birds from the viewpoint of the continuum concept (Curtis & McIntosh 1951) and demonstrated a high correlation between the order of avifaunal similarity and the phytosociological order within the upland deciduous forests of southern Wisconsin. However, several species showed little or no quantitative response to the vegetational differences along the gradient of xeric to mesic sites and the response of each species to this gradient was unique. These classificatory problems are viewed dynamically by Elton (1966) who has proposed a system of habitat inventory which extends beyond the physiognomic and structural characters normally used in the classification of vegetation. It is aimed at a synthesis that will describe not only the parts of a complex system but also the interaction and balance between them, as well as the dynamic properties of the system as a whole (Elton & Miller 1954). Such a classification demands a long-term study of habitats with their varying features and of animal populations in them. On the one hand major requirements of the species must be studied in relation to the principal limiting factors of the environment and the characteristic features or structures of the habitat. On the other hand the dynamic aspect of a species network must be followed through the utilization of resources by various key species. This kind of approach in animal ecology is largely neglected in Australia. This is partly because ecological problems demanding practical solutions are identified in terms of human exploitation of resources in semi-arid environments where physical factors fluctuate violently (Andrewartha & Birch 1954). However, the coastal fringe of the Australian continent supports relatively stable environments of what are often called the zoogeographical 'refuge' areas in which a number of species of birds are considered to have evolved simultaneously (Gentilli 1949; Serventy 1953; Keast 1961). Here the