Reinventing Edmonton's Approach to Zoning Into a Fine Grained Flexible System.

Introduction

Edmonton has a problem. We are entering a phase in our development and planning where we need a much more nimble and effective approach to zoning. The key problems are the following:

• Our land use bylaw is rooted in the 1950s and is inherently inflexible resulting in a proliferation of ‘DC’ zones and a proliferation of ‘standard’ zones, but still it provides almost no flexibility 
• Changing our land use bylaw is a process would be political minefield that would be fraught with difficulty and it would take years. 

I offer a proposed solution that would move Edmonton to a ‘variable’ type of coding that Toronto uses.

We can change the Land Use Bylaw to be a modern and nimble Land Use bylaw allowing us tremendous flexibility for future planning efforts. We can do it in a way that won’t make (many) people angry because we don’t need to change anyone’s land use zoning through its implementation.

 

Toronto’s Land Use Bylaw

Toronto's land use bylaw uses a ‘variable’ code based system for residential areas. So, for example, a specific area might have a zone that reads as follows.

R (f12, a275, u2, d0.55)

In short, this is Residential zone with three ‘variables’ presented. These are a minimum lot frontage (f) of 12 meters, an allowance of 2 dwelling units (2) and a FAR of 0.55.

Essentially this is a variable based system which allows tremendous flexibility in the use of zoning. Using this system, Toronto essentially can have 1000s of ‘zones’, each of which have subtle differences. Thus it becomes possible with a single zone to alter things like the number of units, the permitted height of buildings, various setbacks, maximum site coverage, etc, in a very fine grained way.  Here is a look at the zoning map using this type of code.

Proposal: Reformulate Edmonton’s LUB into a Variable Code format w/o Making Significant Changes to Existing Development Rights.

My proposal is that instead of overhauling Edmonton's zoning bylaw, we simply reformulate it into a Toronto like code based format. If I were to take the standard RF1 zone (non-MNO), it could be done as follows.

                 RF (U1, SAR1, MH 10.0, SCR1, F6, R7.5)

With the following ‘Variables’

“U” Number of Primary Units Permitted
“SAR” Site Area Ratio - A value that can be multiplied by the minimum site area outlined in section 110.4 of the LUB for RF1.
“MH” Maximum Height in Meters
“SCR” Total Site Coverage Ratio - A value that can be multiplied by the Maximum site coverage table in the current RF1 zone.
“F” Front Setback in meters
“R” Rear Setback in meters

This does not mitigate the need for the text for the RF1 zone as the details and exceptions are laid out in it. For a sample of a RF1 zone employing this ‘code’ approach see an attached PDF by clicking here.

Further, the MNO complicates this a bit. However, I do believe it would be possible to integrate many of the elements of the MNO into this system with some work in such a way that the MNO could be significantly reduced in its content (with most of the specifics transferring to the zoning map and the code) and perhaps eliminated altogether. Note, while this is imperfect and would differ from location to location on the zoning map, a sample of the MNO RF1 code could be as follows:

                    RF (U1, SAR1, MH 8.7, SCR1, F3, R7.5)

Note, to truly maintain the exact MNO impacts, the new zoning map would need to be developed through an area by area review to calculate things such as rear and front setback.

Finally, it is worth noting that Toronto’s code leaves out the ‘variable’ letters whenever the standard situation applies. This cleans up the coding and simplifies the map a bit. For clarity, I have included ALL variables in my samples here.

Combining RF1 through RF5

One benefit is that we could combine the RF1 through RF5 into a single zone with nearly all of the differences dealt with through the use of the new zoning code approach and careful mapping of the new zones. A sample of what RF3 would look like would be as follows (again this is non-MNO).

                RF (U3, SAR1, MH 10.0, SCR1, F6, R7.5)

Click here for a PDF draft of the written text for a combined RF1 and RF3 zone that would differentiate sites based upon the code and the specific mapping of sites.

Again, it is important to note that there would be no (or at least minimal) changes to the development rights of any one site by adopting this new code initially. So, a site zoned RF3 prior would be relabeled and the LUB language would change, but the actual development rights would be unchanged. The reason for this is to avoid combining land use changes with the redevelopment of the LUB into this more flexible version.

 

So, if Nothing Changes, Why Do This?

The key benefit, however, is it introduces flexibility in the future to make changes to specific areas. So, if we wanted to reduce setbacks and allow for higher buildings in a four block area, we could do so. If we wanted to allow houses within one block of arterials to be 1 meter taller, with a reduced setback and greater overall lot coverage, we could do so.

Recently, our Mayor suggested that we should change the zoning in Pre WWII areas, this would allow us to easily do that. In many pre-WWI areas (all in MNO) most of the existing houses are above the 8.6M identified in the MNO. The height variable could quickly be changed to ensure that the heights actually align with the local form. In areas where there are existing short setbacks, this could be reflected in the code and on the zoning map.

At present, whenever we want to do something that doesn’t fit within the standard zones, we either have to create a special area zone or a DC1 or DC2 and as such we have a proliferation of them. This complicates our overall LUB.

Simply put, the new zoning approach would reflect our maturity of our city and would give us flexibility to grow and change in the future. 

 

Some Key Points

I don’t know to what level, it would be possible to do this with the current Alberta legislation, so that would need to be checked.

While I have a good knowledge of zoning, I am not an expert in Land Use Bylaws, so I offer this up as an idea and look forward to any feedback and/or insights from others.

Toronto’s LUB includes an ‘x’ code for exceptions. This is a very powerful addition that allows for a lot of details to be inserted into specific areas regarding development regulations and guidelines, etc. I think this is a good idea, but have left that out of this discussion to avoid confusion.

This proposal is the result of about 3 hours of coffee fueled mania, so I have no doubt that it has some errors and inconsistencies in some of the zoning details. A more careful analysis would be beneficial. If anyone is interested and has some funding, I would be happy to put some students on a project furthering the concept.

Fully incorporating the MNO and many location specific elements (i.e. semi-detached only permitted on corner lots, etc) would require a time intensive process of analysis to develop the new zoneing map, but the actual text of the LUB could be simplified with this being done on the map. Alternatively, the MNO could be retained with only some elements dealt with though the ‘variables’

The flexibility of the system depends upon the number of ‘variables’ uses in the zoning text. I have proposed 6 of them. These could be reduced or increased (for example, I have not included side setbacks as a variable. There are benefits and costs to increasing or reducing the number of ‘variables’

I have proposed an approach that emphasizes no changes or very few changes to existing development rights because I think it could speed the adoption of this. There might be merit to allowing for more actual changes to development rights to simplify the resulting LUB zone(s).

Next Steps

First, any feedback would be appreciated. Maybe this is a crazy idea. Let me know by comments or email (Robert.summers@ualberta.ca) or by twitter (@rjscity).

Second, I think it would be best to take a neighbourhood or two and mock this up. It would also be good to play around with different versions. I truly think a combined RF1 through RF5 approach is possible.

Thanks for reading, I look forward to any feedback.