Journal of Housing and the Built Environment 19: 91109, 2004. 2004 Kluwer Academic Publishers.

. Printed in the Netherlands.

Ecological footprints and sustainable urban form

ERLING HOLDEN
Western Norway Research Institute, P.O. Box 163, 6851 Sogndal, Norway (E-mail: erling.holden@vestforsk.no)

Abstract. This paper presents the results of a four-year research project (19972001) entitled Housing as a basis for sustainable consumption. The overall aim was to obtain more empirical and theoretical knowledge about the connection between physical urban planning and household consumption. This knowledge provides a platform for discussing principles and practices for sustainable urban development. This project was based on two main assumptions. First, it was suggested that the signicant and increasing environmental damage due to private household consumption presents a major challenge in achieving sustainable development. Second, a large part of this consumption appears to be inuenced by our physical living situation, i.e., the way we design and locate our houses. This also applies to energy use for heating and technical appliances, transport, and even to the considerable amount of equipment that is needed for household operation, redecoration and maintenance. With respect to transport, the study team included both everyday travel and leisure-time journeys in this research. While everyday trips such as travelling to work, shopping and taking the children to school are strongly inuenced by the living situation of the household, this might also be true for leisure-time travel. Based on two large surveys in the Norwegian towns of Greater Oslo and Frde, the study team collected data on housing-related consumption from 537 households. Ecological Footprinting was then used as an analytical tool to analyse the environmental consequences of this consumption. These ecological footprint analyses suggest that sustainable urban development points towards decentralized concentration, i.e., relatively small cities with a high density and short distances between the houses and public/private services. Key words: decentralized concentration, ecological footprints, planning, sustainable consumption, sustainable development, sustainable urban form

1. Introduction This article is based on a planning research project. More specically, it concerns those areas of planning research that deal with increasing knowledge about the effects of physical planning. Bjrn Re (1990) points out that physical planning should form the basis of the decisions or measures that form our environment and that inuence human activity. One of the basic assumptions in this article is that physical surroundings inuence human behaviour.

92

ERLING HOLDEN

The concept of sustainable development rst appeared on the international agenda around 15 years ago. The UN report entitled Our Common Future (WCED, 1987) pointed out that mankind now faces such major problems with respect to the depletion of natural resources, increased pollution and poverty that something must be done. Unless action is taken to improve these conditions, we risk destroying the planet on which we live and not simply for the current generation; future generations, and even nature itself, are also in the danger zone. Another basic assumption in this article is that we are now faced with a need for change. These two concepts, physical planning and sustainable development, together comprise the general area of research covered in this article. While physical planning constitutes the professional point of departure, sustainable development acts as the thematic boundary. However, this combination provides the basis for a large number of researchable approaches to the matter. Based on the more general concepts, a three-step demarcation was devised in order to arrive at the specic area of research. The rst demarcation relates to a specic aspect of sustainable development, namely consumption. Agenda 21, which is one of several follow-up reports to Our Common Future, states that the most important cause of the steady deterioration in the global environment is todays non-sustainable consumer and production patterns, especially in the industrialized countries. However, production and consumption are closely linked, and it is impossible to imagine the one without the other. The study team therefore chose to focus on consumption. There are three reasons why attention should be drawn to the consumer aspect. Firstly, environmental problems are being increasingly linked to the use of products and services. It is no longer the factory manufacturing products that necessarily presents the most serious threat to nature and the environment. Instead it is the use of these products that gives cause for concern.1 Obviously this should not be taken to mean that it is no longer important to focus on the environmental problems caused by industry and manufacturers, but it represents a shift in the area of interest. Secondly, the team recognize that, under current social conditions, consumption is the real driving force. Non-sustainable production and consumption levels are primarily a result of our desire to do more, experience more, see more, and, to put it briey, consume more. Efforts must be made on the consumer side to lead us onto a sustainable path. Finally, the focus on consumption is based on this articles association with physical planning and the effect that this has on human activity. This demarcation provides the basis for a transition from the general concept of sustainable development to the more specic term sustainable consumption.

ECOLOGICAL FOOTPRINTS AND SUSTAINABLE URBAN FORM

93

The second demarcation focuses attention on a specic part of the combined public and private consumption. Obviously all consumption can be problematic from an environmental point of view. Nevertheless, certain areas of consumption are more relevant than others when talking about physical planning and sustainable development. Housing is a key concept in this context. The type of consumption that can be linked to housing and, of course, to the people who live in these households is particularly relevant. To put it bluntly, developments over the last decade have provided a basis for maintaining that environmental problems have left industry and moved into peoples homes, i.e., housing that has largely been designed and located via physical planning. More specically, four consumer categories are used and referred to as housing-related consumption. The rst is energy consumption with regard to heating and operating housing. This accounts for almost 30% of Norways total energy consumption (Hille, 1995).2 However, housing does not just consume energy. A substantial amount of material housing consumption is required in order to operate and maintain a housing unit, and this is designated as the second consumer category. This concerns furniture and other ttings, technical equipment and electrical appliances, equipment for maintaining and operating indoor and outdoor areas, etc. Individually, these products do not represent major consumption, but together they represent extremely high consumption levels. According to Rolness (1995) just under NOK 30 billion is spent each year on renovation and maintenance alone, with considerable amounts being spent on miscellaneous ttings.3 Perhaps the most prominent feature of our consumer patterns during the last decade is the huge increase that has occurred in the transport sector. From 1951 to 1991 transport energy consumption increased almost sevenfold (Nss, 1997). The home is often the departure point for much of this transport consumption, which includes travelling to work, day-care centres, schools, shops and various leisure activities. These routine journeys are referred to as everyday travel. Energy consumption relating to everyday transport constitutes our third consumer category. A common feature of these three consumer categories is that they are all obviously relevant in the context of physical planning. However, a fourth consumer category has also been included, i.e., energy used for longer holiday and leisure trips. This category has been added to enable investigation of whether the so-called compensation hypothesis can be conrmed. Briey, this hypothesis states that people who expend small amounts of energy on everyday transport (due to certain housing attributes) undertake longer journeys in their leisure time in order to compensate for needs that are not fullled where they live. For example, someone who lives in a densely

94

ERLING HOLDEN

populated urban area with little greenery around them might travel farther aeld in their leisure time in order to reach the wide-open spaces that they dont normally see. If such compensatory effects apply, this could have major consequences for physical planning; e.g., what is the point of continuing to reduce the need for everyday travel if it only results in more extensive travel during holidays and leisure time? The third and nal demarcation concerns various aspects of physical (urban) planning. According to Re (1990), physical planning relates to design at all levels from overall design at a national level down to the design of individual housing. This article deals with four specic planning factors that describe key aspects regarding the design and localization of housing. Physical planning can inuence these planning factors, while at the same time the planning factors affect the extent and composition of housing-related consumption. These four planning factors are: (1) town size/national settlement pattern; (2) localization of houses within a town, municipality or built-up area; (3) residential area; and (4) type of housing. These four factors can be linked to more overriding housing planning principles. The question of a towns size and national settlement patterns is closely related to the question of centralization versus decentralization at a national level. The localization of housing refers to the distance from the house to the centre of town and relates to urban sprawl, while residential areas can also be linked to a discussion about density. Obviously a residential area, where housing is divided into densely populated and sparsely populated areas, is not the only measure of density. In a discussion about density, additional criteria for measuring density should therefore be included in the assessment, e.g., population density and development density. Finally, the question of housing type deals with the ongoing debate about single-family houses as a separate form of living, compared to more dense and concentrated forms of development. Furthermore, these four planning factors are closely inter-related. They inuence each other and, in the overall scheme of residential planning, it can be difcult to consider them as clearly separate aspects.

2. Objectives and issues This article aims to present new knowledge about the relationship that exists between the four planning factors, on the one hand, and housing-related consumption on the other. The issues at hand are:

ECOLOGICAL FOOTPRINTS AND SUSTAINABLE URBAN FORM

95

Do various aspects of the design and localization of a house affect our consumer pattern? What are the overall physical characteristics of a living situation that has the smallest negative impact on the environment? (The term living situation is used here to express the physical design and localization of a house. Impacts on the environment are measured in terms of ecological footprints.) This knowledge is important for two reasons. Firstly, it is interesting to study the extent to which physical-structural conditions affect our actions. Secondly, this is a form of knowledge that can be valuable for specic physical planning. Some of the implications for planning that this revealed knowledge implies are addressed at the end of the article. This includes a brief discussion of the complex issue: What is sustainable urban form? It is important to emphasize here that this article primarily provides a snapshot of the situation in Norway at the end of the 1990s. All results must be evaluated in the light of this transitory and spatial limitation.

3. Methodology The research plan consisted of an empirical part and a theoretical part. The former concerned obtaining new knowledge about the relationship that exists between housing-related consumption and the factors that affect its extent and composition (section 4). The latter concerned incorporating the results of the empirical research into a discussion in the light of other knowledge (section 5). This also involved a discussion as to which principles and criteria ought to be used as a basis for the design and localization of residential areas, within the context of sustainable development objectives. The empirical research plan consisted of three phases, each with an individual approach. Surveys were carried out among a large number of households in two Norwegian cities. These surveys took the form of a questionnaire distributed by post. The aim was to obtain an idea of how housingrelated consumption varied under different living situations. Analysis of this data would also provide an idea as to what percentage of these variations can be linked to the actual living situation, seen in relation to the socioeconomic, sociodemographic and attitude-related characteristics of the individuals concerned. The surveys were carried out between October and November 1998. Case studies4 were also included, to obtain a deeper insight into the mechanisms that inuence peoples consumer habits in complex everyday situations. These studies were designed to provide an understanding of how people experience the effects of physical-structural factors on their choice

96

ERLING HOLDEN

of consumables for their routine everyday lives and leisure time. These case studies were primarily built up around qualitative research interviews carried out within each household. A total of 24 case studies were undertaken during the period April to November 1999. Ecological footprint calculations were made in order to link consumption and sustainable development. These calculations indicated which overall living situations based on the consumer categories highlighted in this article resulted in the least serious environmental consequences. In addition to the data obtained from the survey, these calculations were also based on a quantity of empirical data relating to the environmental consequences of different types of consumption. The focal points in these studies were the households and the types of housing in which the respondents lived. The characteristics of the individuals concerned were also included, to provide a supplementary or alternative perspective. 3.1. Survey The survey formed a basis for describing how consumption varies between different housing types and localities. It consisted of a questionnaire sent to households in Greater Oslo and in the western Norwegian community of Frde. Greater Oslo, which comprises the capital Oslo and the surrounding district, with a population totalling approximately 1 million, represents consumption patterns and volume in a large urban context. Frde, on the other hand, with only around 12,000 inhabitants, gives a corresponding picture for rural conditions. The team carried out a stratied probability sample in order to ensure an adequate number of respondents from different housing types (single-family houses, semi-detached houses and multi-family residential buildings) and housing localities (central/suburban, sparsely/densely developed) within each of the study areas. The distinction between urban and rural areas was ensured by the selection of these two study areas. The questionnaire primarily focused on surveying housing-related consumption and other consumption (mainly consumption in connection with holidays and leisure activities) based on physical and structural conditions concerning the location of the house, as well as attitudes to individual, more general environmental problems. The survey consisted of two separate forms: one completed by the entire household as a group, and the other completed by each individual household member over 18 years of age. Data was collected on the following conditions:

ECOLOGICAL FOOTPRINTS AND SUSTAINABLE URBAN FORM

97

consumer behaviour: information was collected on a broad range of housing-related consumption with regard to conditions (directly or indirectly) connected to the house. Household consumption was also studied in connection with holidays and recreation activities; characteristics of the each house: such as housing type, size (m2 oor area), construction type (wood, brick, concrete) and the total size of the plot (m2); the physical and structural properties of the surroundings: data was collected on inter alia services within walking distance (5001000 m) of the house (shops, public ofces, commercial services, etc.), the distance to the nearest service of each type, as well as the density of buildings in the immediate vicinity and local community; socioeconomic and sociodemographic background data on the individuals living in the households; environmental attitudes: e.g., attitudes to general, environmental political issues.

Figure 1. Survey areas. Greater Oslo (Stor-Oslo), including the municipalities of Oslo, Brum, Asker, Skedsmo, Nittedal, Rlingen, Oppegrd, Ski and Lrenskog. Frde refers to with the municipality of Frde.

98

ERLING HOLDEN

A total of 537 households completed the questionnaire. There seemed to be a reasonable ratio between the sample and the population of each town, regarding the physical characteristics of the houses and a broad spectrum of socioeconomic background factors. It should, however, be mentioned that respondents between the ages of 30 and 60 were slightly over-represented, as were the higher-educated respondents. The number of female respondents was also just above the actual percentage of women living in these areas. 3.2. Ecological footprints The housing-related consumption pattern for each household was translated into an ecological footprint for the household. Before presenting the results of these calculations and the implications they have for planning, let us briey explain the history and concept of ecological footprinting as a tool for environmental impact assessments. The concept of ecological footprinting was developed and quantied by William Rees and Mathis Wackernagel in the early 1990s as an elaboration of the carrying capacity concept (Wackernagel and Rees, 1994). This is a dynamic concept that was changed and improved throughout the 1990s. Numerous books and articles on it have been published, including empirical studies as well as theoretical and methodological publications. One of the latest articles on ecological footprints (Wackernagel et al., 2002, published in the Proceedings of the National Academy of Science 2002), calculates the footprint for the entire world. The concept, although still fairly immature, is now accepted as an important part of the sustainability debate, both by academics and politicians. Ecological footprinting is mentioned as a valuable analytical device (MFA, 2002) in the Norwegian National Strategy for Sustainable Development. So what exactly is an ecological footprint? The basic answer is that we all need a certain amount of land area to survive and that it is possible to calculate this area. Everything that we consume, or dump, needs an area somewhere in the world to produce or assimilate what we use or throw away. As such, ecological footprinting is a simple accounting tool that adds up human impacts (or use of ecological services) in an index, in a way that is consistent with thermodynamic and ecological principles (Chambers et al., 2000). Ecological footprinting is certainly not the only accounting tool around. There is a seemingly innite number of tools on the market, including Life Cycle Analysis, Ecological Space, Ecological Rucksack, Environmental Impact Assessments, Factors 4 and 10, MIPS (material intensity per service) etc. Each of these tools has its advantages and shortcomings, as does ecological footprinting. Before presenting ecological footprinting in more detail, let us briey examine its pros and cons.

ECOLOGICAL FOOTPRINTS AND SUSTAINABLE URBAN FORM

99

There are ve important aspects that make ecological footprint analysis a valuable tool for sustainability analyses. First, the method is based on the life cycle principle, which is a prerequisite for assessing environmentally sustainable development. Second, the method focuses on consumption. One of the main characteristics of todays environmental problems can be related to the unsustainable consumption pattern in the richer part of the world. There can be no doubt whatsoever that issues regarding consumption patterns and volumes must be a central part of sustainable development. The method can be used for consumption at any level, from an individual person up to a country or even the whole global population. Third, the method draws up a synthesis of a large number of different consumption categories as well as environmental consequences in one single analysis. This makes it possible to carry out overall comparisons, and not just limited analyses of specic components or aspects. Fourth, ecological footprinting incorporates equity and global justice into the analyses. Finally, the method has proven to be an excellent tool for illustrating the challenges of sustainable development, for professionals as well as lay people. Ecological footprint analyses are both educational and motivational. However, the system has several shortcomings and limitations. First, some consumption and emission aspects are not included in the analyses. Ecological footprinting only includes consumption and emissions that require land areas, in some form or another (Lewan, 2000). Important environmental issues relating to emissions of heavy metals, persistent organic and nonorganic materials, radioactive substances etc. are therefore not included. Second, doubts have been raised about the land area methodology, especially the CO2 land area (Jrgensen et al., 2002). Finally, probably the most problematic aspect is the idea of aggregating many different land categories into a single number. Under what heading can forests, arable land and built-up areas be subsumed? So far the answer is land productivities; the productivity of different types of land can be determined by referring to the reported yields of various plant and animal produce. Even though this makes it possible to summarize the different land areas, it should be considered what this actually means. It might also be worth mentioning here that ecological footprints say nothing about peoples quality of life, which is a completely different story and needs to be looked at separately. This was a brief introduction to the pros and cons of ecological footprinting. But how does it work? It is customary to operate with six different land categories: cropland, grazing land, forests, shing grounds, energy footprint and built-up land. Cropland includes the area needed to produce all food (grain, fruits, vegetables, etc.) and non-food crops (cereal for animals, cotton, etc.). The global area used as grazing land corresponds to human

100

ERLING HOLDEN

consumption of meat, dairy products and wool derived from livestock that are not crop-fed. The forest footprint refers to the area required to produce forestry products, which are consumed globally, while the shing ground footprint represents the area required to produce the sh and seafood that we consume. The built-up land footprint comprises infrastructure for housing, transportation and industrial production as well as hydroelectric power installations. Finally, the energy footprint refers to the area required to sustain our energy consumption. This encompasses four types of energy (fossil fuels, biomass, nuclear power and hydropower), each with its own methodology for calculating land area. Ecological footprints can be used in several ways. One of the most popular is to calculate the ecological footprint of a nation (or the entire world) and compare this with the available biocapacity5 of that nation (or the world). In other words: comparing the ecological footprint caused by consumption of natural resources with the earths biological capacity to regenerate them. This brings us directly to the very heart of sustainable development. It is clear that the land area is very unevenly distributed between the rich and poor nations (which of course is well documented). An average individual in high-income countries has an ecological footprint of 6.5 ha/year (approximately the size of nine football elds). At the other end of the scale we nd the people living in low-income countries, with an average footprint of 0.8 ha/year (WWF, 2002). These calculations also form the basis of another alarming issue. According to The Living Planet Report 2002 (WWF, 2002), the global ecological footprint covered 13.7 billion hectares in 1999, or 2.3 global hectares per person. This demand on nature can be compared with the earths productive capacity. Approximately 11.4 billion hectares, slightly less than a quarter of the earths surface, are biologically productive, harbouring the bulk of the planets biomass production. The remaining three-quarters, including deserts, ice caps and deep oceans, support comparatively low concentrations of bioproductivity. Still, according to The Living Planet Report, the productive quarter of the biosphere corresponded to an average 1.9 global hectares per person in 1999. Therefore human consumption of natural resources that year overshot the earths biological capacity by around 20%. But ecological footprints can also be used in less pessimistic and sophisticated ways, e.g. as a simple analytical device for comparing the environmental consequences of two households, such as described in this article. Using the information concerning a households housing-related consumption, the research team simply asked: Where are the households with the lowest ecological footprints?

ECOLOGICAL FOOTPRINTS AND SUSTAINABLE URBAN FORM

101

4. Results This section presents the results of the footprint calculations. The calculations were based on data taken from the surveys and the results are presented according to the four planning factors mentioned in the introduction.6 4.1. Size is not important Figure 2 shows the average ecological footprint per household and per household member. Once again, note that the ecological footprints calculated only concern housing-related consumption and not the households total private consumption. Let us look more closely at what these gures mean, starting with the dimension of urban size. Although it is interesting to compare the two survey areas, caution is recommended for several reasons. Greater Oslo and Frde are two complete living, shopping and working areas, but they are so different (in size, extent, and perhaps also culture) that a direct comparison must be treated carefully. Nevertheless, it is interesting to note that the average ecological footprints per household for the two areas are 1.56 ha/year (Frde) and 1.70 ha/year (Greater Oslo). Per household member, these gures are 0.83 ha/year and 0.76 ha/year respectively. This shows that the inhabitants of the small rural town of Frde have an ecological footprint that is 10% less than their urban counterparts in the larger city suburb of Greater Oslo. What causes this? Mainly differences in travel patterns.7 With regard to daily journeys, the Greater Oslo results are favourable. Per household member, the residents of the capital travel 60 km per week, while the corresponding gure for Frdes residents is 98 km. This is mainly because car density is greater in Frde, where 92% of households have access to a car, compared with only 85% in Greater Oslo. However, if we look at the total distance travelled by car throughout the year, and if we now include the long holiday and leisure journeys, this picture is reversed. Despite less car access, households in Greater Oslo have the greatest mobility. In fact, household members in Greater Oslo travel an average of 1,500 kilometres more per year and travel further on privately booked air ights. This implies that average household members in Greater Oslo use 14% more energy each year on private transportation than their rural counterparts in Frde. 4.2. High density, less urban sprawl and less single-family housing Figure 3 shows the ecological footprint per household member according to residential area, distance to the city centre and different types of housing. There can be no doubt that high density, moderate distances between houses

102

ERLING HOLDEN

Figure 2. Average ecological footprint per household in Frde and Greater Oslo. (All numbers in hectare/year.)

and the city centre, and concentrated forms of housing are the most favourable for reducing a households ecological footprint. There are many reasons why dense and concentrated housing turns out positively, from an environmental point of view. First, sparsely populated areas have a much higher percentage of single-family (detached) houses. People living in single-family houses have a signicantly higher energy consumption as well as material housing consumption than people in all other types of housing. Second, the houses are generally larger in sparsely populated areas, which again inuences consumption patterns signicantly. Finally, the percentage of households with access to a private car is higher in sparsely populated areas. Car access is the most important factor in inuencing a households transport energy use. Everything that has been said about densely versus sparsely populated areas also applies to distances to the city centre. Households living near the city centre tend to live in multi-family residential buildings or smaller houses and have less access to their own car than those living near, or on, the urban fringe. One interesting point should be made, however. On average, household income levels are generally higher for those living in densely populated areas and near the city centre than for those living in sparsely populated outlying areas. However, in spite of this additional income, people living in the city centre have a lower ecological footprint.

ECOLOGICAL FOOTPRINTS AND SUSTAINABLE URBAN FORM

103

Figure 3. Average ecological footprint per household and household member in Frde and Greater Oslo according to residential area, distance to the city centre and types of housing. (All numbers in hectare/year.)

When it comes to types of housing, the single-family (detached) house is a poor alternative, at least with regard to the ecological footprint. On average, the ecological footprint per household member is almost 20% higher than for people living in more concentrated types of housing, i.e. semi-detached or terraced houses and multi-family residential buildings (blocks of ats). 4.3. The signicance of non-physical factors of inuence The question that occurs is: Right enough, you nd that physical living situation matters, but is it certain that size, density, distance and housing type are behind these differences? Is it not possible that the differences are really due to other conditions such as social class, income, and the composition of the household? Or at least are they not due to a combination of these factors? Yes, of course conditions such as income and household composition matter. But even when such conditions are controlled for using multivariate regression techniques, the physical/structural dimensions remain central to the households ecological footprint. It should be underlined, however, that we did not nd unambiguous signicant relations between all planning factors and footprint size. Further investigations are therefore needed to verify

104

ERLING HOLDEN

the connections between the living situation and the ecological footprints of the inhabitants. Which non-physical factors play a role? The analyses show three predominant factors with signicant inuence on the ecological footprint per household member. First and most important is the number of people living in the house. There is an economy of scale present where the footprint can be shared among more people. The second factor is car occupancy. Households with access to their own cars have a signicantly higher footprint than those without. The third one is income. The income that households have at their disposal has signicance in both places. The fact that the number of people living in the household, car occupancy and income are important for the size of the ecological footprint comes as no surprise to us. What is interesting in these analyses, however, is that the planning factors also have a strong inuence on the households footprint.

5. What is sustainable urban form? Let us return to our initial research question: What are the overall characteristics of a living situation with the smallest negative impact on the environment, i.e., the smallest ecological footprint? We are looking for an environment-friendly living situation that helps reduce a households housing-related consumption as much as possible. This is also a living situation that allows us to avoid any compensatory effects, e.g., in the form of long holidays and leisure trips. Based on the material obtained from the survey and the calculation of the households ecological footprints, four attributes in the housing situation seem to produce the best results in reducing the ecological footprint. These are: dense and concentrated housing design; relatively high degree of density in residential areas; shortest possible distance to the town centre; moderate size of location. But what about the issue of sustainable urban form? What implications do the norms for sustainable development have on the design and localization of houses? According to Nss (1997), there are two competing models of sustainable urban development. On the one side there are those who support compact cities. The idea here is that large, dense and concentrated cities will support the principles of sustainable development. However, on the other side, there are those who support the green city, i.e., a more open type of urban structure, where buildings, agricultural elds and other green areas form a sort of mosaic-like pattern.

ECOLOGICAL FOOTPRINTS AND SUSTAINABLE URBAN FORM

105

The supporters of the compact city8 believe that this has environmental and energy advantages, as well as social benets. The list of advantages is remarkably long, including a better environment, affordable public transport, potential for improving the social mix, and a higher quality of life (see Frey, 1999, pp. 2125 for a supplementary list). However, the main justication for the compact city is the need to promote the least energy-intensive patterns of activity to help us cope with the issues of global warming (Frey, 1999). But opponents insist that the case for compact cities is not proven because this concept fails to acknowledge the poor prospects for reversing deep-seated decentralization trends (Breheny, 1992). The list of arguments against the compact city is even longer than that of points in its favour; the compact city implies the rejection of suburban and semi-rural living, neglect of rural communities, less green and open space, increased congestion, increased segregation, and less power for making local decisions (see Frey, 1999, p. 25 for a supplementary list). Until fairly recently there was some consensus that compact urban forms (i.e., the compact city) offered the most sustainable future (Williams et al., 2000). Although there has always been considerable scepticism, the concept of a compact city is so dominant that it seems inconceivable that anyone would oppose the current tide of opinion towards promoting greater sustainable development and the compact city in particular (Smyth, 1996, p. 103). In this context, it is not surprising that the move towards the compact city is now entrenched in policy throughout Europe (Jenks et al., 1996, p. 275). Our research also strongly supports the idea of the compact city. However, the important aspect of urban size still needs to be considered. In the compact city concept, two different pairs of concepts are often mixed together without further qualication. These are centralization-decentralization and concentration-sprawl (Hyer, 2002). The former refers to the population patterns in larger national contexts, the latter to the development processes within urban areas. Since the early 1960s and the advent of the car era, urban development may be characterized as centralized sprawl. This means centralization of the overall national population pattern, and sprawl of each of the urban concentrations. In some cases researchers have concluded that the concept of a compact city implies further centralization of the population pattern and that larger cities are favourable in a sustainable urban development (Newman and Kenworthy, 1989; CEC, 1990). Our research does not support such claims. On the contrary, more favourable ecological footprints result from having several smaller compact towns and cities, rather than a few large or mega cities. This is termed decentralized concentration, and it opposes the dominating development patterns of the last decades in every aspect (see Figure 4).

106

ERLING HOLDEN

Figure 4. Four models for sustainable urban form.

It is, however, a fact that an increasingly larger percentage of the population both in Norway and in the rest of the world live in large cities. This is the situation that urban planners have to face. In a modern democracy people cannot be transferred from large cities to smaller and more compact towns and cities. Not even in the name of sustainability. The answer to this challenge is therefore to encourage polycentric cities, which implies dense and concentrated centres within the large cities. These centres should contain a variety of housing and workplaces, as well as private and public services. It is also vital that these polycentric cities are built on an effective and environmentally sound public transport infrastructure that connects the different centres.

6. Conclusions and nal remarks This article shows that decentralized concentration could lead to smaller ecological footprints of households a conclusion that seems to be enjoying widespread support (Breheny, 1992; Bannister, 1992; Owens, 1992; Newman and Kenworthy, 2000; Buxton, 2000; Masnavi, 2000; Hyer and Holden, 2001). This could be integrated into a policy that strengthens the existence of smaller compact town and cities throughout the country, or into one that encourages decentralized concentration within existing cities. According to Breheny (1992), the concept of decentralized concentration is based on sustainable development and urban form policies such as slowing down the decentralization process and realizing that compact city proposals, in any extreme form, are unrealistic and undesirable. As such, various forms of decentralized concentration, based around single cities or group of towns, may be appropriate. Furthermore, inner cities must be reju-

ECOLOGICAL FOOTPRINTS AND SUSTAINABLE URBAN FORM

107

venated and public transport must be improved both between and within all towns. People-intensive activities must be developed around public transport nodes, along the lines of the Dutch principle of the right business in the right place. This implies that mixed use must be encouraged in cities and zoning discouraged. Finally, urban (or regional) greening must be promoted and combined heat and power (CHP) systems must be promoted in new and existing developments. Such profound changes will take a long time to achieve, as Frey (1999) underlines. We fully concur with Breheny (1992, p. 22) that the real challenge is . . . to redesign existing urban form. Some important elements can be changed quickly (e.g., bus routes), but other elements, such as railway networks and commercial buildings, can only be changed infrequently. Acknowledgements This article is based on a research project that was implemented in cooperation with the Norwegian Institute for Urban and Regional Research (NIBR). The study team are particularly indebted to the research carried out by Ragnhild Skogheim at this institute and Karl Georg Hyer at Western Norway Research Institute (WNRI). We are also indebted to the Research Council of Norway and their research programme on Sustainable Production and Consumption, which made the project possible. Notes
1 The private car is a good example of this. Approximately 90% of the energy used by a car

throughout the total life cycle can be attributed to the use phase (i.e., driving). Only a small portion (less than 10%) is related to the production (and possible demolition) of the car. 2 Applicable to both direct and indirect energy consumption. 3 We must emphasize here that, as far as the material housing consumption category is concerned, we have concentrated only on the type of consumption that relates to running a house or apartment. Material consumption, with respect to new construction and demolition work, has not been included. 4 The results of the case studies are not included in this article. However, they were a part of the overall research plan and are therefore mentioned briey here. 5 Biocapacity (or biological capacity) refers to the total biological production capacity per year of a biological productive space, for example inside a country. It can be expressed in global hectares. 6 The specic survey data from each of the four consumption categories will not be given here. The data that are briey mentioned in the text are presented fully elsewhere (Holden, 2001; Hyer and Holden, 2001). 7 The average ecological footprint per household member for energy consumption in the home and material housing consumption were both about equal for Frde and Greater Oslo.

108

ERLING HOLDEN

The differing circumstances point us in different directions but, in total, the two come out fairly equal (Holden, 2001). 8 Including CEC (1990), Jacobs (1961), Newman and Kenworthy (1989), Elkin et al. (1991), Scherlock (1991), Enwicht (1992), McLaren (1992).

References
Bannister, D. (1992) Energy Use, Transport and Settlement Patterns. In: Sustainable Development and Urban Form (Ed, Breheny, M.), Pion, London. Breheny, M. (1992) Sustainable Development and Urban Form: An Introduction, In: Sustainable Development and Urban Form (Ed, Breheny, M.), Pion, London. Buxton, M. (2000) Energy, Transport and Urban Form in Australia. In: Achieving Sustainable Urban Form (Eds, Williams, K., Burton, E. and Jenks, M.), E&FN Spon, London and New York. CEC (1990) Green Paper on the Urban Environment. Commission of the European Communities, European Commission, Brussels. Chambers, N., Simmons, C. and Wackernagel, M. (2000) Sharing Natures Interest. Ecological Footprints as an Indicator of Sustainability, Earthscan Publications Ltd. Elkin, T., McLaren, D. and Hillman, M. (1991) Reviving the City: Towards Sustainable Urban Development, Friends of the Earth, London. Enwicht, D. (1992) Towards an Eco-City: Calming the Trafc, Envirobook, Sydney. Frey, H. (1999) Designing the City. Towards a More Sustainable Form, E&FN Spon. Hille, J. (1995) Sustainable Norway, The Project for an Alternative Future, Oslo. Holden, E. (2001) Boligen som grunnlag for brekraftig forbruk (Housing as basis for a sustainable consumption), Doktor ingeniravhandling (Ph.D. Dissertation), Department of Town and Regional Planning, Norwegian University of Science and Technology, Trondheim, p. 115. Hyer, K.G. (2002) Analyseverkty i miljplanleggingen Verkty for mer enn festlige anledninger? (Analytic tools in environmental planning). In: Fra miljvern til brekraftig utvikling i kommunene. Erfaringer med Lokal Agenda 21 (From environmental protection to sustainable development in municipalities. Experiences with Local Agenda 21) (Eds, Aall, C., Hyer, K.G. and Lafferty, W.), Gyldendal Akademisk, Oslo. Hyer, K.G. and Holden, E. (2001) Housing as Basis for Sustainable Consumption, International Journal on Sustainable Development, 4(1), 4858. Jacobs, J. (1961) The Death and Life of Great American Cities. The Failure of Town Planning, Random House, New York. Jenks, M. Burton, E. and Williams, K. (Eds) (1996) The Compact City: A Sustainable Urban Form? E&FN Spon. Jrgensen, A.E. et al. (2002) Assessing the Ecological Footprint. A Look at the WWFs Living Planet Report 2002, Environmental Assessment Institute, Kbenhavn. Lewan, L. (2000) Ecologiska fotavtryck & biokapacitet verktyg fr planering och uppfljning av hllbar utveckling i ett internationellt perspektiv (Ecological footprints and bio-capacity tools for achieving sustainable development), The National Board for Housing, Building and Planning in Sweden, Karlskrona / Swedish Environmental Protection Agency, Stockholm. McLaren, D. (1992) Compact or Dispersed? Dilution is No Solution, Built Environment, 18(4), 268284.

ECOLOGICAL FOOTPRINTS AND SUSTAINABLE URBAN FORM

109

Masnavi, M-R. (2000) The New Millennium and the New Urban Paradigm: The Compact City in Practice. In: Achieving Sustainable Urban Form (Eds, Williams, K., Burton, E. and Jenks, M.), E&FN Spon, London and New York. MFA (2002) National Strategy for Sustainable Development, Ministry of Foreign Affairs, Oslo. Newman, P. and Kenworthy, J. (1989) Gasoline Consumption and Cities. A Comparison of U.S. Cities with a Global Survey, Journal of American Planning Association, 55(1), 2437. Newman, P. and Kenworthy, J. (2000) Sustainable Urban Form: The Big Picture. In: Achieving Sustainable Urban Form (Eds, Williams, K., Burton, E. and Jenks, M.), E&FN Spon, London and New York. Nss, P. (1997) Fysisk planlegging og energibruk (Physical Planning and Energy Use), Tano Aschehoug, Oslo. Owens, S. (1992) Energy, Environmental Sustainability and Land Use Planning. In: Sustainable Development and Urban Form (Ed, Breheny, M.J.), Pion Limited, London. Rolness, K. (1995) Med smak skal hjemmet bygges (The home should be built with taste), Aschehoug, Oslo. Re, B. (1990) Fysisk planlegging faglig innhold og utfordringer konsekvenser for utdanningen (Physical planning content and challenges consequences for education). Forum for utdanning i Samfunnsplanlegging. Konferanse 1213 Juni 1990, Norges Tekniske Hgskole, Trondheim (The Norwegian Institute of Technology, Trondheim). Scherlock, H. (1991) Cities are Good for Us, Paladin, London. Smyth, H. (1996) Running the Gauntlet: A Compact City within a Doughnut of Decay. In: The Compact City: A Sustainable Urban Form? (Eds, Jenks, M., Burton, E. and Williams, K.), E&FN Spon. Wackernagel, M. and Rees, W. (1994) Ecological Footprints and Appropriated Carrying Capacity, Island Press. Wackernagel, M. et al. (2002) Tracking the Ecological Overshoot of the Human Economy, Proceedings of the National Academy of Science, pp. 16. WCED (1987) Our Common Future, The World Commission on Environment and Development, Oxford University Press. Williams, K., Burton, E. and Jenks, M. (Eds) (2000) Achieving Sustainable Urban Form, E&FN Spon, London and New York. WWF (2002) Living Planet Report 2002, World Wide Fund for Nature, Gland, Switzerland.