PII: S0006-3207(97)00076- 1 Biological Conservation 84 (1998) 65-81 © 1998 Elsevier Science Ltd All fights reserved. Printed in Great Britain 0006-3207/98 $19.00 + 0.00 ELSEVIER DEFORESTATION, URBANISATION AND SEASONALITY: INTERACTING EFFECTS ON A REGIONAL BIRD ASSEMBLAGE Carla P. Catterall, Mark B. Kingston, Katherine Park & Sven Sewell Faculty of Environmental Sciences, GriJfith University, Nathan, Queensland4111, Australia (Received 17 June 1996; revised version received 2 April 1997; accepted 7 April 1997) Abstract We assessed the effects of conversion of eucalypt forests and woodlands (bushland) to cleared or developed land on bird communities in the rapidly urbanising lowlands of south east Queensland, in subtropical Australia. The majority of 56 species analysed showed large and signifi- cant density differences between bushland and cleared land: 22 (39%) were classed as 'bushland species', 13 (23%) as 'developed land species', and 21(38%) as "generalists' or 'inconsistent'. 'Bushland" species showed a strong correspondence with those identified in other stu- dies as exhibiting regional declines since European settle- ment, whereas 'developed land' species corresponded with those exhibiting regional expansions. Comparisons o f summer and winter abundance also showed large differ- ences in many species, mainly because of seasonal migra- tion. Lowland eucalypt forest remnants of this study were characterised by (1) the presence of a suite of distinctive common bird species different from those which occur in cleared and developed areas, and (2) substantially higher total densities in winter than in summer, due mainly to bushland-dependent winter immigrants. The latter are at risk of further declines with ongoing habitat loss. In con- trast, bird communities of the developed areas showed few seasonal differences. © 1998 Elsevier Science Ltd. All rights reserved. Keywords: birds, deforestation, seasonal migration, subtropics, fragmentation. INTRODUCTION During the past two hundred years there has been widespread change to wildlife habitats within the Aus- tralian continent. For example Graetz et al. (1995) esti- mated that native vegetation cover has been cleared from about 20% of the land area, and significantly dis- turbed over about 50%. Forests and woodlands have been particularly affected by clearing, and may have been reduced in cover by 40-60% on a continental scale (RAC, 1991). The pattern of deforestation has not been 65 uniform across the continent, and some regions have lost 90% or more of their former forest or woodland cover (for example, the Western Australian wheatbelt, Saunders et al., 1993a; and parts of the Murray-Darling basin, Goldney and Bowie, 1990). Historical and contemporary records of bird species have been used to show that large concurrent changes in regional and continental bird assemblages have accom- partied this loss of forest cover. Some species have increased in range and/or abundance while others have decreased (see for example Blakers et al., 1984; Saunders and Curry, 1990; Saunders, 1989, 1993a, b). Many of these changes in bird abundance and distribution may be attributed to the interaction between deforestation and bird species' habitat requirements, where certain species can maintain populations only in areas of intact native vegetation, whereas others achieve greatest abundance in cleared areas. The latter should be also those which are expanding geographically, and the former should be declining. Clearance of habitat was the most frequently inferred cause of decline within Garnett's (1992) review of Australia's 150 threatened and extinct bird taxa. Categorisation of bird species into 'response guilds' according to their proneness to being affected by human impacts on their habitat has been advocated by Croon- quist and Brooks (1991, 1993). However these and other authors have used mainly subjective judgements and qualitative inferences in allocating species to categories. For example a high noted or remembered number of sightings of some species within cleared areas, and a restriction of sightings of other species to wooded areas, has been used as an adjunct to the interpretation of historical regional changes in abundance (Blakers et al., 1984; Saunders and Curry, 1990; Saunders, 1989, 1993a), or to place species in categories of dependence on natural habitat prior to other analyses (e.g. Loyn, 1985, 1987; Barrett et al., 1994). The limited literature providing formal quantitative descriptions of the effects of conversion of forested land to cleared land contrasts with a recent proliferation of research into the effects of fragmentation on popula- tions and communities of birds and other wildlife,