| Assessing the impacts of sprawling urban economies
Retail and industrial parks have become a familiar sight on the outskirts of European cities. Whether it’s Amsterdam, Madrid, Bristol or Tel Aviv, employment density in the centre is declining, and activities that generate jobs are increasing on the fringes. The process is known as deconcentration – more specifically non-residential or economic deconcentration – because it describes not the spread of housing but the spread of commercial activities which provide jobs and services.
|Carefully planned commercial development, 15km outside Prague.|
What effect does this type of sprawl have on our quality of life? Can it be managed for the benefit of the citizens? SELMA (Spatial Deconcentration of Economic Land Use and Quality of Life in European Metropolitan Areas) is a three-year project analysing the effects of economic deconcentration on quality of life. It aims to help governments and planners see how different policies relating to urban development can impact on people’s lives, through changes to their social, physical and economic environment. The project ends in March 2006. It has produced a handbook on the effectiveness of different policy instruments for controlling urban sprawl, and has also successfully used a computer model called UrbanSim to simulate the urban environment in three European metropolitan areas.
City centres of attention
SELMA involves ten partners from seven countries. Most are research institutes focused on urban development and planning. Partners from each country studied two of their own cities, one large, one smaller, to provide information for the project from 14 cities in all.
The research first defined patterns of sprawl using four key attributes: the magnitude of deconcentration, whether it is concentrated or diffuse, the type of development that characterises it (industrial, service or retail) and whether the policy context is market- or planning-led. Each city was given a profile according to these attributes. Bristol, Southampton and Madrid, for example, all show a large amount of deconcentration, with multiple, widely scattered centres of development. Their new development is largely in the service sector, which generally represents the latter stages of the deconcentration process. Employment in the outer rings of the British cities has grown by up to five times the national rate over the last 15 years.
By contrast, Pescara-Chieti in Italy and Prague and Brno in the Czech Republic show small amounts of deconcentration, focused in certain areas, either due to the influence of infrastructure or the historical effect of planning policy. It is generally commercial rather than service sector development. In most countries, urban sprawl is led by the demands of the market, with the exception of Denmark and Holland, where planning regulations have genuine control over development at local, regional and national levels. This ‘typology’ of urban deconcentration is a vital starting point in understanding the processes that influence urban development.
The next stage was to define ‘quality of life’ indicators relevant to the effects of urban deconcentration. Seventeen indicators were chosen to represent different facets of city life in each case-study city. They included, for example, the average speed of rush hour traffic, the percentage of dwellings exposed to high noise levels, the proportion of land area housing buildings, the spending per capita per local authority, and the number of jobs available.
The project then examined how specific policies in different countries have affected the pattern of deconcentration and the quality of life indicators. One might expect US-style market-led governance, with little control over urban development, to lead to greater deconcentration than social-democratic style governance where urban sprawl is heavily regulated. “Our biggest surprise was finding that these contrasting approaches do not systematically affect urban sprawl,” said project coordinator Martin Dijst of the University of Utrecht. “For example, in Italy there are hardly any policies to influence deconcentration, yet there is little severe urban sprawl around cities.” The project has defined a number of best practice approaches to controlling urban sprawl, such as maintaining an urban boundary like a ‘green belt’ and planning a city on the basis of public transport accessibility, as in Copenhagen.
|The development of downtown Rotterdam.|
The final stage of SELMA was to develop a computer simulation of economic deconcentration and its effects on quality of life indicators, so the impact of various policy options could be evaluated. The partners chose an American model, know as UrbanSim, developed at the University of Washington and freely available on the internet. “We had several land-use models already in the Netherlands,” explained Professor Dijst, “but this open source programme is easy for other countries to access.”
In fact, adapting UrbanSim to the European context was not that easy. Tuning the model to the available datasets was time-consuming. Since there is no straightforward user interface, computer programmers were needed to adjust the model for each city. However, the project has set up the model to provide an output for many of the SELMA quality of life indicators for three cities – Tel Aviv, Rome and the Randstad Northwing Region of the Netherlands, which includes Amsterdam and Utrecht. These models provide an extremely useful tool, allowing each city to plan for urban development that improves rather than degrades the lives of its citizens. “We hope our work will be an inspiration to other countries,” concluded Dijst.
Following on from SELMA, the Dutch partners, Utrecht University and the Netherlands Environmental Assessment Agency, are already helping another country, Portugal, to develop the UrbanSim model for some of its cities.