Backyard bonanza: improving the quality of ‘popular’ suburban infill

Abstract

The subdivision of suburban lots to accommodate additional dwellings has been one of the most popular mechanisms for achieving urban infill in Australian cities. This is certainly the case in Perth where the subdivision of individually owned residential lots still accounts for approximately 80% of all infill dwellings created. These ‘successful’ piecemeal intensification projects are permitted through the up-zoning of existing suburban areas and encouraged through changes to strata subdivision legislation and increasing land values. They are evidence of some alignment between the supply and demand sides of the infill housing equation – a necessary, and yet somewhat elusive condition in today's economic landscape. Local governments are, however, somewhat reluctant to continue to facilitate this process of residential intensification, despite sustained pressure to increase infill dwelling numbers, citing poor outcomes in terms of the resulting qualities of the residential environment as a chief reason. How could the current residential planning and design policy (R-codes) be successfully amended to re-enable a pre-existing and demonstrably popular mechanism, which potentially aligns both demand and supply, and thereby significantly contribute to meeting infill targets in the future? The research, undertaken in partnership with the City of Fremantle, attempts to address this question.

Keywords:

• urban infill
• greyfields
• urban design
• residential environment
• subdivision

ORCID

Anthony Duckworth-Smith http://orcid.org/0000-0003-1966-7211

Notes

1 Sprawl is simply defined as an excessive consumption of land within a metropolitan area to accommodate growing population (Brueckner 2000), and often characterised as a low-density phenomenon either in terms of population or dwelling numbers. Other metrics consider in more detail the form and land use structure of urban development. Population density is a reasonable measure of comparing the degree of sprawl amongst cities whose characteristics of urban form and land use structure are similar (Ewing, Pendall, and Chen 2003;Huang, Lu, and Sellers 2007).

2 For the purpose of this study, suburban areas are defined through lot structure and dwelling density – individually owned lots which are predominantly exclusively residential with standalone dwellings accessed via private vehicle driveways off streets, with a maximum net density of 30 dwellings per hectare and a typical gross density of 10–15 dwellings per hectare.

3 The residential environment can be generally defined as the territory concerned with the human experience of inhabiting the built environment relating to shelter and dwelling (Rapoport 1969). It is a complex multi-dimensional concept and comprises both physical attributes (equipment, facilities and services) and social dimensions (friendships, collective behaviours and casual relations). It can be understood as a series of socio-spatial domains that define particular groupings of behaviours and actions (Rapoport 1977). This model of relatively discrete domains is often both explicitly and implicitly cited in much discussion relating to the residential environment (Amérigo and Aragonés 1997, Manzo and Perkins 2006) and generally identifies at least the ‘dwelling’, ‘neighbourhood’, ‘community’ and ‘city’ as the principal socio-spatial territories. The neighbourhood domain is sometimes broken down further into a micro-neighbourhood, which reflects a more familiar, communal environment of predominantly indirectly related acquaintances.