Context: Conservation of biodiversity in cities depends on ecologically sensitive urban planning, informed by an understanding of patterns of species distributions and richness. Because urbanized landscapes are heterogeneous mosaics, and many species move between different land-cover types, it is valuable to compare ‘whole landscapes’ (broad-scale spatially heterogeneous areas) that systematically differ in landscape structure. Objectives: We tested the relative influence of housing cover and canopy tree cover on avian species richness, to identify the components of landscape structure that most strongly influence landscape-scale richness (i.e., the pooled richness of multiple sites within a whole landscape). Methods: We selected 30 residential landscapes (each 1 km2) in Melbourne, Australia, stratified to represent concurrent gradients of housing and canopy tree cover. Five point-count surveys were conducted at each of 10 sites per landscape (for a total of 50 surveys per landscape) and the data pooled to represent the whole landscape mosaic. Results: Up to 82% of variation in avian richness was explained by properties of the whole landscape. Housing cover was most dominant and a strong predictor for multiple response groups including native, terrestrial, forest, and aquatic birds. As housing cover increased, the richness of all groups decreased. Tree cover, primarily comprised of scattered trees in residential areas, had less influence on richness. Nonetheless, for forest birds, the extent of native vegetation surrounding a landscape had an important positive influence, indicating the value of potential source habitat for urban bird populations. Conclusions: Cities can be home to a diverse avifauna. The strong influence of landscape structure on species richness indicates a scope to plan and manage urbanized areas to support a diversity of birds that require natural habitat elements. We conclude that urbanizing environments can best be designed to benefit native birds by protecting patches of native vegetation (particularly large source areas) combined with localized higher housing cover, rather than uniform (lower) housing cover across the entire landscape.
Funding
This study was funded by the Holsworth Wildlife Research Endowment and the Ecological Society of Australia, BirdLife Australia, the Australian Wildlife Society, and the Field Naturalists Club of Victoria. JEH was supported by an Australian Government Research Training Program (RTP) Stipend Scholarship.