The other day I posted visualisations of what fully deemed to comply ResCode envelopes would look like, out of concern for the lack of any such imagery provided by government ahead of the new provisions becoming operational on 31 March. The day before yesterday (the Friday before the provisions come into effect) the government posted a new guidance document, the Town House and Low Rise Guidelines. This is essentially an update to the old Understanding the Residential Development Standards Practice Note, and provides assistance in understanding the provisions.
The new material does not include the diagrams of deemed to comply envelopes that I believe should be provided, but it did include another diagram I wanted to see – an illustration of the 10% tree canopy requirement. This is at page 35 of the document and reproduced above.
The tree canopy requirement has been highlighted as a new requirement in the code in the government’s announcements, and it is true that there was no numerical standard for tree canopy cover in the old provisions – this was instead part of a more qualitative assessment of landscaping. The centrality of the messaging about canopy makes sense, as a commitment to protecting and enhancing tree canopy is a key action in the new Plan for Victoria (see Action 12). This one element in that strategy that has carried through relatively strongly from Plan Melbourne 2017-2050, which had also included strong language about urban greening and cooling.
I agree this should be a priority – as climate change makes our cities hotter, adequate tree canopy is going to be ever more important. But do the provisions match the rhetoric?
The key question is whether the new approach to tree canopy is a step forward or back, and whether it will help achieve that Plan for Victoria‘s target of 30% tree canopy coverage. This is why I was so eager to see some visualisation or testing of the outcomes of the new tree canopy standard.
How did it work until now?
The previous ResCode landscaping provisions are reproduced below. (Apologies to those reading on their phone or on other devices where these screenshots don’t reproduce well!)
These don’t set a quantitative numerical standard for the amount of landscaping or tree canopy cover. Instead the decision-maker has to be satisfied that the first four points (the Objectives) are met, informed by the points in the Standard and decision guidelines. That was a qualitative judgement, which would be informed by the site context, existing vegetation, and the landscape plan provided wiht the application.
Notably, that includes several points encouraging retention of mature vegetation. In particular:
- The decision-maker has to be satisfied that the fourth objective relating to encouraging mature vegetation has been met.
- The Standard specifies development should provide for retention of trees (albeit linked to character outcomes, rather than other values of trees); this even extends to replacement of trees removed in the preceding 12 months.
- It is valid to consider, in assessing the landscape plan, whether a tree was removed to gain a development advantage. This opens up the door to require equivalent landscaping – and by implication, preventing development of part of the site – if the decision-maker considers a tree was inappropriately removed to improve development yield.
These provisions effectively acted as a “soft” tree removal control. Even where no permit was required to remove trees, the provisions helped to deincentivise tree removal to facilitate the development. This was a response to the rightly-hated practice of “moonscaping” – stripping a site of all vegetation to make redevelopment easier. In my view the controls were pretty successful at that aim – over more than two decades of assessing and working on ResCode applications, I can think of only a few examples where I felt developers had tried to remove high quality trees before a permit process occurred.
Was the lack of detailed quantifiable standards a problem? Well, certainly appropriate landscape responses was probably one of the most common areas of dispute between councils and permit applicants. On the other hand hand, I would argue that the desirability of responding to existing vegetation is one of the key clearest examples of why the system should be requiring contextual responses to individual sites. Planning processes should be targeted at areas where they add value, and I think this is such a case.
Another key issue with these provisions that is worth noting, before turning to the new controls, is that vegetation was seen almost entirely through the lens of neighbourhood and landscape character. Habitat is mentioned, but the objective actually only allows assessment of this in “locations of habitat importance.” What that meant was not clearly explained, and a decision-maker would need to justify that at least some level of elevated level of habitat importance existed. The provisions also did not mention urban greening and cooling.
As a result of this, assessment of landscaping and tree removal in medium density housing assessment was very focussed on character. This, in my view, reinforced unfortunate inequities in how planning processes treat different neighbourhoods: trees were valued more in leafy affluent suburbs, where there was a strong landscape character, than in more disadvantaged areas that frequently have fewer trees.
This meant that the provisions were not as clear on the intrinsic value of trees, both new and especially existing trees, for habitat and urban cooling.
The new provisions
The new provisions are shown below.
These provisions are now purely about tree canopy – all references to other aspects of landscaping have been dropped. Crucially, they are deemed to comply, which means that in the first instance, only the Standard is assessed. If the 10% (or 20% on larger sites) tree canopy is provided, subject to the other criteria such as building separation and soil volumes, there is no ability to make any further requirement. (Unlike the old provisions, the decision guidelines cannot be considered in such a situation.)
Finally, I note that the new decision guidelines do now mention greening and cooling. However as mentioned those decision guidelines only come into play if the Standard is not met. Habitat value of urban trees is alluded to in the objective’s mention of biodiversity- but again, this will not come into play if the standard 10% cover is met.
This is why it is so critical that the 10% figure gives good enough outcomes, and why I wanted to see what the state government feels those 10% canopy outcomes look like.
What does 10% canopy cover look like?
As I said at the outset, we now have a little more detail in the Town House and Low Rise Guidance document. This document’s tree canopy diagram is reproduced again below, this time with the explanatory notes.
(An aside before I talk about the tree and landscaping issues: the development yield in the diagram is ridiculously low – especially for the 860sqm lot on the left. This diagram – and the various other diagrams in the guidelines showing low yield dual occupancy development – further reinforces my concern that the government has not properly worked through the typologies that will appear under the code. I also want to make it clear that I am not saying we need to live in a world where medium density development looks like the above, but with more trees. There is ample scope, in my view, for a good design code to increase residential densities substantially while still getting good canopy outcomes).
Some questions
These diagrams still leave me with a lot of questions.
Is 10% canopy a good outcome?
This is the biggie, obviously. I was very sceptical of the 10% figure when announced, but despite this was still genuinely shocked when I saw the diagram above. This is, in my view, a really poor tree canopy outcome. This is especially the case when you consider the dropping of other landscape requirements (more about this below).
My own view is that the tree canopy to be achieved under the new provisions is likely to be less – and potentially much less – than was typically achieved under the old provisions in most suburban scenarios. (It might lead to increased canopy on inner suburban development where gardens were typically smaller). The landscaping above would, in my experience, have been unlikely to be considered sufficient in most situations under the old provisions.
Could I be wrong? Sure. But this is where we come back to justification and explanation of the controls. Some testing could have been done to assess typical canopy and landscaping outcomes currently being achieved, to help benchmark then new provisions. What percentage of tree canopy is typically shown on current landscape plans? If that work has been done, it hasn’t been released. It should be.
Does 10% canopy align with Plan for Victoria’s canopy target?
Action 12 of Plan for Victoria states that the government will “increase the tree canopy cover of urban areas by setting a target of 30 per cent tree canopy cover.” Having a 10% canopy cover standard on private land is not necessarily at odds with this objective – a great deal of canopy will be expected to be provided on public land such as parks and road reserves. But we don’t really have any clarity about what that breakdown looks like. To achieve 30% cover across urban areas, what sort of coverage do we need to achieve on private land? And how does that translate to targets on new development? Was work done to ensure that the 10% standard and 30% target aligned?
Perhaps working this out would have been prohibitively complicated. However, even if this is the case, the strategy clearly acknowledges that the 30% target involves an increase of tree canopy from current practice. This takes me back to the site-level analysis suggested above. If we need an increase in tree canopy, we should be able to have some idea of whether the new provisions are likely to result in an increase on private land compared to the old provisions.
No support of that proposition has been provided, and as I say above, it seems unlikely to me.
Do the provisions enable and even incentivise wholesale removal of existing trees?
As noted, the provisions remove the “soft” protection for existing trees that was in the old provisions. They also create a situation where compliance will be much easier to demonstrate by relying on proposed trees – which exist as precise circles on a pan – rather than the vagaries of measuring an existing canopy.
Another issue is that existing trees cannot count toward the standard if they are within 4m of an existing building. New trees are not subject to this restriction. It is also, of course, much easier to construct a building without having to worry about protecting existing trees on site.
It therefore seems to me the provisions make it far preferable for developers to use new trees to achieve the standard than to rely on existing trees.
The controls therefore seem to me likely to encourage a return of the old style moonscaping, which as I said the old provisions had been reasonably effective at eliminating. This is a huge backward step in terms of retaining and protecting urban canopy.
(Incidentally, Plan for Victoria includes a passing reference to a tree removal control applying to trees over 5m tall “in many cases” but there’s no sign of that in these provisions.).
What other landscaping can be required?
As mentioned, the provisions remove the old landscaping objective, and also removes council’s discretion to seek more than the standard specifies. This seems to mean that council cannot require any other landscaping than the tree canopy. This is where the canopy and landscape outcome shown in the diagrams above goes from poor to genuinely diabolical. Those few measly trees are all the plants you are going to get under a minimally compliant design.
I say “seems to mean” above because I have been told that government representatives have disputed this in information sessions with councils, suggesting a broader discretion exists to require landscaping other than the minimum tree canopy. But I cannot see how that can be the case. The General Residential Zone, for example, has a decision guideline requiring consideration of “proposed landscaping” for non-residential use and development, but this is not included for dwellings. The provisions pointedly turn off council’s broader discretion, and indeed the whole point of these changes is that meeting the new code assures approval.
If the government really believes landscaping beyond the minimum tree canopy requirement is or can be required, it should clarify the basis for this as a matter of urgency. However the simple fact is that the code removes other landscaping requirements, and the government assures us that development meeting the standard will be approved. It can’t have it both ways.
Are the controls workable?
This is a larger critique of the whole codified ResCode project – about which I have already said a lot – but the provisions are in my view far too complex to provide the system efficiencies that are used to justify them. Codification shifts ResCode Standards from discretionary assessment guidance into process-critical, black letter law process rules. Councils need to be certain throughout the processing of an application whether or not the development complies with every last point of the provisions, as the legal rights of neighbours, and the council’s procedural obligations, hinge on compliance. Mistakes, or simply areas of doubt, will lead to legal disputes and delays, undermining the supposed process improvements.
Given all this, process rules need to be much simpler than these provisions to be workable.
10% of site area is a relatively simple metric, although this measurement is complicated by its application to overlapping circles. While the actual area calculation can presumably be largely automated, the provisions above show a lot of points of potential error and dispute. Trees can be moved in and out of the calculation based on the amount of soil they are sitting, a judgement about species (which incidentally is contrary to the founding premise that the controls are now wholly quantitative), and how close they are to buildings. There are a lot of moving parts.
There are also going to be some areas that I expect will be subject to a lot of dispute. One immediate challenge is trees overhanging boundaries. The diagram above seems to suggest that canopy outside the site can be counted towards a site canopy total as long as the trunk of the tree is on the land. That makes some sense, in principle, but raises all sorts of practical problems. Can canopy not contained on the land actually be relied upon to delivered, given neighbours could prune such canopy? Can developers line their trees along the boundary to minimise the amount of canopy on their land? Does this interfere with the development potential of the neighbouring site, given they presumably cannot double count this canopy for their own development? And so on.
That is before we consider existing trees. As already mentioned, trees are not perfectly round and symmetrical, with precisely measurable canopies. When existing trees are being relied upon, there will be much more scope for dispute about what the area of actual canopy is. As I have already suggested, I think the complexities around existing trees will unfortunately incentivise their removal.
These points might seem like nitpicking, but this is exactly where this kind of system design – using very complex codes as process rules – takes us. The design of the code – and this is across all of the provisions – is too complex to be used as a process critical deemed to comply mechanism. This is poor regulatory design. I doubt that we will see improved timelines from these provisions.
Final Thoughts
It is good that we at least have some visualisation of a compliant outcome for tree canopy. This is more than we can say for the ResCode building envelopes, as I said the other day. But the images we have seem to show a really dire landscape and tree canopy outcome. In both cases, we have very little to go on to understand the justification and testing done on the controls.
If the tree canopy and landscape outcomes are better than I believe, and will help increase canopy cover, why can’t we see the testing or justification that supports this?