Animal Conservation (2005) 8, 309–319 C  2005 The Zoological Society of London. Printed in the United Kingdom doi:10.1017/S1367943005002258 Inferring persistence of indigenous mammals in response to urbanisation Rodney van der Ree † and Michael A. McCarthy Australian Research Centre for Urban Ecology, Royal Botanic Gardens Melbourne, c/o School of Botany, University of Melbourne, 3010, Victoria, Australia (Received 15 September 2004; accepted 16 December 2004) Abstract The number and extent of human settlements around the world are increasing as the human population grows and people move from rural to urban areas. The decline and ultimately the local extinction of many indigenous species from a range of taxonomic groups within urbanised areas is a recurring theme from cities around the world. However, determining whether a species has become extinct can be difficult because sufficient survey effort may be too costly and expert opinion may involve considerable subjective bias. In this study, we compared the results of four quantitative assessments of the probability of persistence of indigenous mammals within Melbourne, Australia. Our comparisons were methodological, taxonomic and spatial. Less than half (26) of the original 54 species that occurred in Melbourne prior to European settlement have a greater than 95% probability of being extant at the end of 2000. Mammals occurring in local government areas (LGAs) within 10 km of the central business district of Melbourne were less likely to be extant, with 29% of species having a greater than 95% probability of persisting, compared to 48% in the outer LGAs. The group of species most negatively affected by urbanisation were the small, ground-dwelling mammals, with just two out of 15 species having a greater than 10% probability of persisting. All four methods gave broadly similar results, with the Bayesian approach consistently suggesting higher probabilities of species persistence. The decline in the number of species is likely to continue with the ongoing expansion of Melbourne. The greatest opportunity to conserve the maximum number of species of mammal within Melbourne is in the outer LGAs where they remain extant. We recommend that state and local governments design and adopt a comprehensive strategy for managing habitat networks that cross jurisdictional boundaries and encompass the greater Melbourne area to enhance prospects for the survival of mammals. INTRODUCTION Human-dominated and human-impacted ecosystems can now be considered the norm rather than the exception (Vitousek et al., 1997). The greatest alterations to ecosystems have occurred in landscapes where natural habitats have been completely removed and replaced by different land-uses. This transformation is evident in urban areas and where industries have been established to supply goods and services to urban populations, such as agriculture, forestry and mining. The proportion of the world’s human population residing in urban environments is predicted to increase to 66% by 2025 from 9% in 1900 (see McIntyre, Knowles-Yanez & Hope, 2000). The continued growth in both the number of people contained within cities and the spatial extent of city boundaries will exacerbate the loss of biological diversity within cities as well as in the areas that service the needs of human inhabitants. † All correspondence to: R. van der Ree. E-mail: rvdr@unimelb.edu.au The major cities of Australia have grown dramatically in size since their relatively recent establishment in the years after European settlement of the continent in 1788. This pattern of establishment and growth means that for some cities, it may be possible to determine the composition and distribution of some taxonomic groups prior to habitat loss and modification. In contrast, the date of establishment of many cities and towns, particularly in Europe, Asia and Africa may extend back many centuries or in some cases millennia. Consequently, our ability to fully document the ‘before-development’ levels of biodiversity in these old cities is limited by the absence of detailed information on the distribution or abundance of species. However, while it may be possible to compile a list of species known or likely to have occurred in relatively young cities, determining their current status may be problematic. Quantitative methods that utilise atlas or herbaria data to infer the probability that a species still persists have recently been developed (Solow, 1993a,b; Burgman, Grimson & Ferson, 1995; McCarthy, 1998; Hedenas et al., 2002; Solow & Roberts, 2003). Similar statistical methods have also been used to calculate that the actual date of