Do “brownfields” sites matter?

I was leafing through Challenge Melbourne, the discussion paper released in 2001 as part of the Melbourne 2030 process, the other day. This very interesting but apparently long-forgotten factoid caught my attention:

“Capacity for an estimated 65,000 dwellings on large sites such as old factories has been identified in the established suburbs”.

Given that the number of households in Melbourne is projected in Victoria in Future to grow by 825,000 between 2006 and 2036, it seems the potential contribution from “brownfields” sites – mainly large disused industrial and public sector sites – will be modest.

Melbourne @ 5 Million envisages that 53% of the required new dwellings will be located within the established suburbs. If the 65,000 figure is even broadly close to the mark, it seems that the great bulk of this new housing will have to come from redevelopment of small sites, most of which are presumably residential and likely to generate significant opposition from neighbours.

The key issue this raises is whether or not the anticipated level of redevelopment in established suburbs is achievable. Brownfields sites have made a significant contribution over the last 20 years to construction of multi unit housing but apparently will make a relatively small contribution in the future.

It is possible the Challenge Melbourne figure has been superseded. Unfortunately Melbourne 2030 does not mention a comparable figure. It just indicates that 255,000 dwellings are expected to be accommodated on “strategic redevelopment sites (particularly Principal Activity Centres and Major Activity Centres) within established metropolitan urban areas”. There is nothing on what sort of sites they might be. Nor does Melbourne @ 5 Million indicate a figure (in fact it devotes more space and real substance to new fringe development than to the more politically sensitive issue of redevelopment within established suburbs)

What I would like to know is how reasonable were the assumptions used in Challenge Melbourne to estimate the likely dwelling yield from brownfields sites. Perhaps 65,000 was very conservative (or perhaps not). Did the methodology take full account of likely resident opposition and constraints like heritage issues and residual industrial contamination? How many of the sites identified in Challenge Melbourne have already been developed and has the rate of development exceeded the identification of additional sites?

In short, has the “low hanging fruit” largely been accounted for already?