Thursday 18 June 2020

Urban Design and the RMA: Urban Planning v Urban Design

Further thoughts on urban design under the RMA.

It is important to distinguish urban design effects from other effects. Otherwise there may be double counting. For example, there is a big overlap between urban planning and urban design. Are urban design effects just urban amenity effects? There is also overlap with landscape and visual assessment, In this blog, I want to look at urban planning and urban design. 

There are different ways to conceptualize this overlap:

Urban design has been described as second order design. First order is buildings or spaces. Third order is (spatial) planning. Urban design is said to be the link in between these two orders, finding first order designs that help fill in the third order outcomes. Some urban designers will baulk at this distinction, as urban design works across all spatial scales.

Another way to look at the issue is that urban design follows a different process to urban planning. Urban planning follows a structured process of identifying outcomes, options analysis, making trade offs and selection of appropriate set of tools to achieve the desired outcome.  Urban planning attempts to be rationale. It is by its nature synoptic (it tends towards the large scale). The end product tends to be a bunch of processes and procedures against which individual developments are judged.

Urban design in contrast could be said to be more place-based and more iterative in its approach to bringing together a range of outcomes for an area or site.   The design process can be a bit irrational and improvised. There may be no universal sequence of stages – analysis, design and implementation are contemporary and inextricably linked. The end product can be a resolved 'plan' for an area - the masterplan or structure plan.

This process type distinction may be valid, but it doesn't really help with distinguishing urban planning from urban design effects. The idea that urban design may operate at a different 'level' to that of urban planning is more useful. But how to describe this in terms of RMA effects?

To  start with, urban design can be said to concentrate on the quality and functionality of the built environment. It is clearly not about management of the natural environment, or about infrastructure planning, for example, although these matters are relevant matters. Urban planning could be said to sit above and span across these different 'inputs'.  Moving on from that, clearly urban planning has a focus on urban quality, but is perhaps more directed at the basic building blocks of urban quality, rather than place-based responses. 

Urban design brings in a strong element of the human response to new and modified urban environments. Perhaps this is the strongest distinction between urban planning and urban design - urban planning needs to take a long term (intergenerational) view of the management of urban resources; urban design is more about the day-to-day response and interplay between people and the built environment.   

Matthew Carmona has suggested in Place Value: Place Quality and its Impact on Health, Social, Economic and Environmental Outcomes (see note 1) that it is possible to envision different urban qualities as sitting on a ladder that climbs from:
  1. those place qualities to be avoided when shaping the built environment (because of their very possible negative impact on place value); 
  2. those place qualities which are fundamental and which should be required as a means to maximize place value through good design.
  3. place qualities that are strongly associated with the delivery of place derived value of all types (and which should be the aspiration of built environment policy and development-related decision making). 
I think the list of things to avoid and those that are essential (number 1 and 2 in the list above) are more to do with urban planning, rather than urban design.

Starting with the qualities that are fundamental:
  • greenness in the built the environment (notably the presence of trees and grass, water, and open space ‒ the latter if of good quality); 
  • a mix of uses (notably the diversity of land uses within a neighbourhood); 
  • low levels of traffic; 
  • the walkability and bikeability of places (derived from their strategic street-based connectivity and the quality of the local public realm); 
  • the use of more compact (less sprawling and fragmented) patterns of development; and 
  • ready convenient connection to a good public transport network. 

And the qualities to be "avoided":
  • car dependent and extensive forms of suburbanization;
  • relentlessly hard urban space (absence of local green space);
  • too much very local permeability;
  • the presence of rear parking courts and other segregated areas;
  • poor maintenance / dilapidation (including of green spaces);
  • a sense of overcrowding in residential areas;
  • the presence of unfavourable food stores; and
  • the impact of roads with higher traffic loads and speeds, wider carriageway widths, or which are elevated.
Im not quite sure what unfavourable food stores are (too may fast food outlets?)  while some of the effects are more operational rather than layout and design (such as poor maintenance).

Urban design can then come more strongly into play in the third sets of qualities, as described as follows:

Next there is a strong positive association between place derived value of all types and fifteen often less tangible, sometimes subjective, and generally more difficult to measure qualities of place. Whilst the evidence on each of these remains powerful, it is not definitive, in the same way as it is for the qualities already discussed. Partly this seems to be because the more ‘difficult’ nature of these qualities makes researching them more challenging, and so there is often less research available on which to make a definitive assessment. There are also greater challenges in specifying exactly what quality means in these areas, making the evidence that is available more equivocal.

These qualities include:
  • visual permeability;
  • sense of place (distinctiveness);
  • pedestrian scale (of streets and buildings);
  • façade continuity;
  • natural surveillance (the creation of defensible space);
  • presence of street level activity / background movement;
  • good street lighting;
  • a denser street network (urban grain);
  • low traffic speeds;
  • low neighbourhood noise;
  • presence of attractive / welcoming / comfortable / adaptable public spaces;
  • positive (sociable) public/private threshold features;
  • integration of built heritage;
  • integration of natural features and a diverse ecosystem; and
  • perceived architectural quality and beauty generally in the built environment.
Carmona notes that whilst some of these qualities, for example façade continuity or traffic speeds, are relatively easily specified, most need more careful interpretation. In his words, they are therefore likely to be ‘aspirational’ rather than required qualities. In RMA terms they may be qualities that need assessment as to the degree to which, if they are missing or only partially applied in a development, adverse  'spillover' effects may arise.  

Interestingly, he has a set of qualities that are not so certain as to the link to place quality. These are:
  • different architectural styles (about which the evidence is simply unclear).
  • higher versus lower densities of development (where within the health research, and with regard to sociability versus perceived crime, the evidence conflicts).
  • extreme densities (where conflicting evidence is apparent relating to carbon reduction, social welfare and ecological richness).
  • high-rise living (where the evidence is unclear, although tending to warn against families living in such circumstances).
  • street length and pedestrian connectivity (where divergences are apparent within the evidence on health versus crime).
  • cul-de-sacs (where, within the evidence on crime and safety and with regard to property value, sociability and children’s play, conflicts are apparent).
  • vehicle / pedestrian separation (about which the evidence is weak and indecisive).
  • use of shared spaces (where conflicts are apparent, particularly with regard to the evidence on actual and perceived safety).
  • the economic impact of the proximity of retail to residential properties (about which conflicts exist on the relative size and impact of negative externalities sometimes associated with local retail)
Some of these qualities are relevant to urban design principles. I guess what is important is that these sets of qualities need to be applied with a high degree of judgement. 



Note 1: Place value: place quality and its impact on health, social, economic and environmental outcomes. Matthew Carmona. Sourced from  https://www.tandfonline.com/doi/full/10.1080/13574809.2018.1472523



Thursday 4 June 2020

Journey to work and housing density



In  theory, there should be a relationship between transport accessibility and urban form. As transport costs increase, cities tend to build up; as transport costs fall, cities tend to spread out. As cities get more densely developed, then high capacity public transport becomes more important in getting people around.

What about Auckland?

Starting with transport, if I plot distance from the CBD (as the crow flies) versus the percentage of bus, train, ferry, walk and cycle trips to work (in 2018) based on home residence, then I get the following graph.

This looks quite a good relationship: as distance from the central area increases, then the number of bus, train, ferry, walk and cycle (BTFWC) trips to work decrease. This data is based on census area units and 2018 census data.



Using the formula provided by Excel, I get the following broad pattern.


Distance (kms) from centre
5
10
15
20
% work trips by BTFWC
27.60
18.25
12.79
8.91

Now lets look at housing density. The following graph plots gross dwelling density (again by Area Unit) by distance from CBD.




The dwelling density figure is a ‘gross’ figure - the land area of the Area Unit divided by the number of dwellings. A low gross density may therefore be because of a large park being present in the Area Unit, not because housing is spread out.

Again, based on the graph, I get the following broad pattern.


Distance (kms) from Centre
5
10
15
20
Dwelling density (dwell/na) 
19.5
11.6
7.0
3.7


There is not necessarily a one to one relationship between the two sets of figures (journey to work and density) but the curves are not dissimilar.

Going from a dwelling density of 11 dwellings per ha to 20 dwellings per hectare will not automatically mean that public transport, walking and cycling will go from  18 to 27% of work trips, as proximity to the centre cleary has a big influence on both density and use of public transport, walking and cycling.

If we look at urban form, rather than density, then the percentage of dwellings that are in an attached form (rather than stand alone) should also increase as distance reduces (and public transport, walking and cycling increases).

The following graph plots the percentage of attached dwellings by Area Unit, based on distance from the centre.




Bit more of a scattered picture than for dwelling density.


Distance (kms) from Centre
5
10
15
20
% Attached
28.9%
21.2%
16.7%
13.5%



If we put the three sets of figures together then we get the following pattern:


Distance from CBD (kms)
5
10
15
20
% BTFWC
27.6%
18.2%
12.8 %
8.9%
dwelling density
19.5
11.6
7.0
3.7
% of dwellings attached
29%
21%
17%
13%

So as we go from 15 to 10kms out from the CBD, the % of BTFWC trips to work increases by 1.43, housing density goes up 1.66 and percent attached dwellings go up by 1.27. 



15-10 kms
20-15 kms
% BTFWC
1.43
1.44
Density
1.66
1.89
% attached
1.27
1.24

It would seem that the regional land use pattern is responding to transport costs and benefits. The question then becomes whether the observed step up in density and number of attached dwellings is responsive enough to the changed transport conditions. Should there be more , or less of a density step up?