Dr Morten Gjerde, a senior lecturer in urban design in the School of Architecture at Victoria University, asks if the Unitec target of 3000-4000 houses can be met without creating a future slum of 40-square-metre dog-box houses. His answer: certainly yes if we look for inspiration from overseas models. [This article first appeared on the Newsroom website and is published with the permission of the author.]
OPINION: Housing Minister Phil Twyford’s recent announcement that the Government plans to build between 3,000 and 4,000 new homes on a site in Auckland’s Mt Albert had many people reaching for their calculators. Could it be right, and is it even possible, that a new residential neighbourhood could be built to a density of between 100 and 133 units a hectare?
When these figures are compared with suburban areas we’re more familiar with, sitting somewhere between 15 and 25 units a hectare, or housing classified as medium density, ranging between 33 and 66, it’s no wonder the critics were queueing.
The question most people are asking is, can these targets be met without creating a future slum of 40-square-metre dog-box houses?
Although that would be one possible outcome, it can easily be avoided through good planning and a willingness to look beyond the standard housing solutions we’ve become used to. Indeed, with the right input of creative thinking and financial resources, the Mt Albert KiwiBuild site could become a highly desirable place to live.
News of the project coincides with that of another Auckland development, the Daisy Apartments, where the 33 apartments have just two car parking spaces to share between them.
In general, developers resist moving away from their standard practices, pitched at what they perceive the market demands. In most cases, this means residential projects will include at least one car parking space per dwelling. After doing his homework, the developer of the Daisy Apartments believes people might be willing to give up their cars if reasonable alternatives such as ride-sharing phone apps and public transport services are available. If he’s right, and there is every reason to believe he is, everyone will win because fewer cars stored on site means more space for people.
Like Daisy, the KiwiBuild site is well linked to the city and region by public transport routes running along its eastern and western sides. Twyford has indicated the area would also have access to high-speed bus services and light rail before the project is completed. This should give the site’s designers licence to think about organising it around people’s needs rather than cars. That is not to say the development won’t be planned with cars in mind, but there will be less need to store cars around the site and the roadways could have narrower carriageways with consequently wider footpaths.
The key to unlocking the potential of this site is for the designers to think beyond the range of housing types commonly used in large-scale developments. While the compact suburbia approach has been implemented effectively in Auckland at Hobsonville Point, it won’t cut it with the KiwiBuild site, because it simply isn’t efficient enough. Apartment blocks stack housing vertically, making more efficient use of the land, but residents can feel isolated in tall towers and miss out on ground-level outdoor living space.
In meeting the Government’s ambitions for the site, designers should give consideration to the perimeter block, a housing format seen extensively throughout European cities.
Perimeter blocks are one of the oldest urban settlement types and can create highly liveable, low-rise housing at relatively high densities.
The key to an effective perimeter block is to plan the area comprehensively and in much more detail than would be needed for standard detached housing.
Perimeter blocks take the form of conjoined terraces, varying between nine and 14 metres in depth and two to four storeys in height. Along the ‘public’ frontage, the houses are normally set back from the street edge to allow for small garden spaces, which creates opportunities for personalisation and a more pleasant street environment.
As individual houses in the perimeter block are tightly spaced they must be laid out to strict controls to make sure everyone can have equal access to sunlight, views and privacy. This housing also works best around the concept of shared amenity, where small, separated areas within a development are combined into a larger, more useful one.
The best example of a shared amenity in this form of housing is to be found in the centre of the block, which can be large enough to provide for significant planting and biodiversity. When created as a shared space, as is common in Europe, the centre can be used for active recreation as well as providing residents with a pleasant outlook from their houses. Where car numbers within the development can be reduced, this central area can be made very people friendly.
Although perimeter block formats can help achieve the Government’s density targets while providing residents with high quality housing, it would be wrong to focus entirely on this option. All communities should offer a range of options, even if the range is constrained by the land available, to enable people to match their housing to lifestyle preferences and family needs. It is important that the KiwiBuild site includes options for stand-alone homes as well as apartment buildings and that the mix of housing types should be anticipated and provided for in the way the site is planned.
Social infrastructure like schools, places of work and services such as shops will also need to be considered. Much already exists, which distinguishes the site from development in outlying areas, and mixed use will help fill in the gaps.
For the Mt Albert Kiwibuild site to fulfil its potential will require a commitment to innovate and to hold fast to the principles of diversity, quality and sustainability. But it won’t be necessary to reinvent the wheel, as the models for living at higher densities can be found in countries with which we like to compare ourselves.
[Editor’s note: The Housing Minister has said that 63 per cent of the 29 hectares will be devoted to residential housing. That’s around 18 hectares – giving a density rate of about 165 units per hectare for a 3000-unit development and a rate of 220 for a 4000-unit site . However, that 63 per cent is unsettled and it may be higher, perhaps quite a bit higher.
When this was raised with Dr Gjerde, he said a lesser figure would of course make the resultant density that much higher. “We would naturally expect there to be variations in relative density across the site and on a site as large as 29 hectares there should be opportunities for meaningful open space. The complicating factor is whether there are buildings on those parts of the site that will not be built on and what effect that will have on perceived density. Whatever the case may be, I believe that good design with a willingness to create liveable neighbourhoods will be necessary. I wouldn’t expect the nett site density to be able to be much above 150 du/hectare to meet those two objectives.”]