sign in sign up
<<
Home , 1998 in Sweden, Ecological economics, Urban ecology, Urban ecosystem

Ecological 29 (1999) 293–301

ANALYSIS

Ecosystem services in urban areas

Per Bolund a, Sven Hunhammar a,b,*

a En6ironmental Strategies Research Group, Natural Management, Department of Systems , Stockholm Uni6ersity, Stockholm, b Stockholm En6ironment Institute, Stockholm, Sweden

Abstract

Humanity is increasingly urban, but continues to depend on for its survival. are dependent on the beyond the limits, but also benefit from internal urban ecosystems. The aim of this paper is to analyze the services generated by ecosystems within the . ‘Ecosystem services’ refers to the benefits populations derive from ecosystems. Seven different urban ecosystems have been identified: street trees; lawns/; urban ; cultivated ; wetlands; lakes/sea; and streams. These systems generate a range of ecosystem services. In this paper, six local and direct services relevant for Stockholm are addressed: air filtration, micro regulation, noise reduction, rainwater drainage, treatment, and recreational and cultural values. It is concluded that the locally generated ecosystem services have a substantial impact on the quality-of- in urban areas and should be addressed in land-use . © 1999 Science B.V. All rights reserved.

Keywords: Ecosystem; Ecosystem services; Urban areas

1. Introduction pected to live in cities (UN, 1997). But even if humanity is increasingly urban, we are still as Humanity is rapidly urbanizing, and by 2030 dependent on Nature as before. Cities are, for more than 60% of the world population is ex- example, dependent on the large hinterlands needed to provide input and take care of from the city. In a study of the 29 largest cities in the Baltic Sea region, it was estimated that the * Corresponding author. fms, Box 2142, 103 14 Stockholm, Sweden. Tel.: +46-8-4023808; fax: +46-8-4023801. cities claimed ecosystem support areas at least E-mail address: [email protected] (S. Hun- 500–1000 times larger than the area of the cities hammar) themselves (Folke et al., 1997).

0921-8009/99/$ - see front matter © 1999 Elsevier Science B.V. All rights reserved. PII: S0921-8009(99)00013-0 294 P. Bolund, S. Hunhammar / 29 (1999) 293–301

When humanity is considered a part of nature, We begin with a general discussion of urban cities themselves can be regarded as a global ecosystems and their ecosystem services. A num- network of ecosystems. If compared with true, ber of local and direct services relevant for Stock- natural ecosystems, the man-made ones are how- holm are then discussed. The paper is concluded ever immature due to features like their rapid by a synthesis and a discussion on the conse- growth and inefficient use of such as quences for . and (Haughton and Hunter, 1994). Odum (1971) even observes cities to be ‘‘only parasites in the ’’. 2. Urban ecosystems But there is also a presence of natural ecosys- tems within the . As will be discussed in An ecosystem can be defined as ‘‘a set of inter- this paper, the natural urban ecosystems con- acting species and their local, non-biological envi- tribute to and increase the quality- ronment functioning together to sustain life’’ of-life of urban citizens, e.g. improve air quality (Moll and Petit, 1994). However, the borders and reduce noise. Most of the problems present in between different ecosystems are often diffuse. In urban areas are locally generated, such as those the case of the urban environment, it is both due to traffic. Often the most effective, and in possible to define the city as one single ecosystem some cases the only, way to deal with these local or to see the city as composed of several individ- problems is through local solutions. In this re- ual ecosystems, e.g. parks and lakes (Rebele, 1994). For simplicity, we have chosen to use the spect, the urban ecosystems are vital. term urban ecosystems for all natural green and The aim of this paper is to analyse some of the blue areas in the city, including in this definition ecosystem services generated by urban ecosystems street trees and ponds. In reality, street trees are and discuss their importance for the quality of too small to be considered ecosystems in their urban life. The emphasis is to identify the services own right, and should rather be regarded as ele- and whenever possible also quantify and ments of a larger system. them, with greatest relevance to cities in We identify seven different urban ecosystems and North America. Examples will be taken from which we call natural, even if almost all areas in the city of Stockholm in Sweden. cities are manipulated and managed by man. The It is difficult to generalize a discussion like the ecosystems are street trees, lawns/parks, urban one in this paper to reflect the importance of forests, cultivated land, wetlands, lakes/sea, and ecosystem services in all cities of the world. Both streams. the actual and its value are site-specific Street trees are stand-alone trees, often sur- and can vary significantly around the world. Cit- rounded by paved ground. Lawns/parks are man- ies differ, since they are built in all kinds of aged green areas with a mixture of grass, larger , their sizes vary from small towns to huge trees, and other . Areas such as playgrounds megacities, and the wealth of city inhabitants and golf courses are also included in this group. ranges from extreme poverty to excessive luxury. Urban forests are less managed areas with a more Methodologically, the identification and valua- dense tree stand than parks. Cultivated land and tion of ecosystem services could be viewed as an gardens are used for growing various items. input to a cost-benefit analysis (CBA) aiming at Wetlands consist of various types of marshes and more efficient land-use in urban areas. The swamps. Lakes/sea includes the open water areas benefits of ecosystems are often neglected in ordi- while streams refers to flowing water. Other areas nary CBAs and if increased values (both mone- within the city, such as dumps and abandoned tary and non-monetary) could be allocated to backyards, may also contain significant popula- ecosystems, the results of CBAs on new in- tions of plants and . It should be possible, frastructure or conservation projects could however, to place most urban ecosystems or ele- change. ments in one of the above mentioned categories. P. Bolund, S. Hunhammar / Ecological Economics 29 (1999) 293–301 295

Our classification is crude and has to be adopted generated services relevant for Stockholm. From to site-specific conditions. the 17 groups of services listed by Costanza et al. Stockholm has a large and varied ecological (1997), six are considered to have a major impor- structure. In the City of Stockholm, parks and tance in urban areas: air filtering (gas regulation), green space occupy 56 km2 (26%), and water areas micro-climate regulation, noise reduction (distur- cover 28 km2 (13%) of the total area of 215 km2 bance regulation), rainwater drainage (water regu- (Miljo¨fo¨rvaltningen, 1995). This is considerably lation), (waste treatment), and more water and green space than possessed by recreational/cultural values. Other services, such most other cities, and gives Stockholm its unique as food and control, could character. The city is situated on a number of also have been included, but are not considered islands between the lake Ma¨laren and significant for Stockholm. For each of the ad- the brackish Baltic Sea. Stockholm also has a dressed services the following aspects are special feature with a number of green wedges discussed: pointing towards the . This allows the “ Which kind of problem does the service con- ecosystems close to the city centre to be linked tribute to the solution of? with larger ecosystems outside of the city. The “ What ecosystems are involved in the generation City of Stockholm has about 700 000 inhabitants. of the service, and how? Greater Stockholm has 1.5 million inhabitants. “ Quantification and valuation of the service with examples from the literature. “ Examples from Stockholm. 3. Locally generated ecosystem services 3.1. Air filtering Ecosystem services are defined as ‘‘the benefits human populations derive, directly or indirectly, Air caused by transportation and from ecosystem functions’’ by Costanza et al. heating of buildings, among other things, is a (1997) and they also identify 17 major categories major environmental and public health problem of ecosystem services. A number of these ecologi- in cities. cal services are not consumed by directly, It is clear that vegetation reduces , but are needed to sustain the ecosystems them- but to what level seems to depend on the local selves. Such indirect services include of situation (Svensson and Eliasson, 1997). The re- plants and nutrient cycling, but the classification duction is primarily caused by vegetation filtering is not obvious. Another aspect of ecosystem ser- pollution and particulates from the air. Filtering vices is that they have different spatial cover. capacity increases with more leaf area, and is thus Services can be available on the local or global higher for trees than bushes or grassland (Givoni, scale according to the scope of the problem they 1991). Due to the larger total surface area of are connected to and the possibility of transfer- needles, coniferous trees have a larger filtering ring the service from where it is produced to the capacity than trees with deciduous leaves (Stolt, city where humans benefit from it. Such a transfer 1982). This capacity is also greater because the can take place both by man-made and needles are not shed during the winter, when the by natural means (e.g. atmospheric transport). air quality is usually worst. However, coniferous Easily transferred services with a global scope, trees are sensitive to air pollution and deciduous like CO2 sequestering, do not necessarily have to trees are better at absorbing gases (Stolt, 1982). A be produced close to the source of the problem. mix of species therefore seems to be the best Services which are impossible to transfer must, alternative. In general, vegetation is much better however, be generated close to where they are than water or open spaces for filtering the air. consumed (e.g. noise reduction). The location and structure of vegetation is im- Since this paper focuses on issues relevant for portant for the ability to filter the air. Bernatzky urban areas, the attention is on direct and locally (1983) reports that up to 85% of air pollution in a 296 P. Bolund, S. Hunhammar / Ecological Economics 29 (1999) 293–301 can be filtered out, and in a street with trees, tioning substantially in urban areas by shading up to 70%. Thick vegetation may simply cause houses in summer and reducing wind speed in turbulence in the air while a thinner cover may let winter. the air through and filter it (Bernatzky, 1983). In Chicago it has been shown that an increase According to some estimates (Tolly, 1988; Bram- in tree cover by 10%, or planting about three trees ryd and Fransman, 1993), 1 ha of mixed per building lot, could reduce the total energy for can remove 15 t of particulates per year from the heating and cooling by US$50–90 per dwelling air while a pure spruce forest may filter two or unit per year. The present value of long-term three times as much. The trees of the Chicago benefits by the trees was found to be more than region have been estimated to remove some 5500 twice the present value of costs (McPherson et al., t of air , providing more than US$9 1997). million of air quality during 1 year (McPherson et The micro-climate in Stockholm is regulated to al., 1997). a great extent by the large bodies of water in the In Stockholm the percentage of vegetated area, city, as the city is situated on a number of islands. as well as of water area, is clearly above the Mean annual are reported to be European average (Eurostat, 1995). In fact, ap- 0.6°C higher in Stockholm as com- proximately 10% (22 km2) of the land area in the pared to areas outside the central city (Alexan- City of Stockholm is forested. Such a large dersson et al., 1991). Stockholm also benefits from amount of forest has a significant air filtering the vegetation, for example by reduced heating capacity which leads to an improvement of air costs. quality. The total filtering service of Stockholm vegetation has not been estimated. 3.3. Noise reduction 3.2. Micro-climate regulation, at street and city le6el Noise from traffic and other sources creates health problems for people in urban areas. The Local climate and even weather are affected by overall costs of noise have been estimated to be in the city. In studies of US cities, some of these the range of 0.2 –2% of GDP in the EU (Kom- differences have been quantified, and expressed as munfo¨rbundet, 1998). In Sweden, maximum noise changes compared with surrounding country-side: levels of 55 dB(A) outside and 30 dB(A) inside air is 0.7°C higher measured as the buildings have been established as the long-term annual mean, solar radiation is reduced by up to goal (Naturva˚rdsverket, 1996). 20%, and wind speed is lowered by 10–30% The distance to the source of the noise is one (Haughton and Hunter, 1994). The phenomenon, key factor, and a doubling of the distance de- sometimes called the urban island effect, is creases the equivalent level by 3 dB(A). Another caused by the large area of heat absorbing sur- key factor is the character of the ground. A soft faces, in combination with high amounts of en- lawn, rather than a concrete pavement, decreases ergy use in cities. the level by another 3 dB(A) (SOU, 1993). Vege- All natural ecosystems in urban areas will help tation also contributes to the decrease, but at to reduce these differences. Water areas in the city what level is uncertain. One source states that a will help even out temperature deviations both dense shrubbery, at least 5 m wide can reduce during summer and winter. Vegetation is also noise levels by 2 dB(A) and that a 50-m wide important. A single large tree can transpire 450 l can lower noise levels by 3–6 dB(A) of water per day. This consumes 1000 MJ of heat (Naturva˚rdsverket, 1996). Another source claims energy to drive the evaporation process. In this that 100 m of dense vegetation is only reported to way city trees can lower summer temperatures of decrease noise by 1–2 dB(A) (Kommunfo¨rbundet, the city markedly (Hough, 1989). Vegetation can 1998). Sounds propagate long distances on water also decrease energy use for heating and air condi- (Naturva˚rdsverket, 1996). P. Bolund, S. Hunhammar / Ecological Economics 29 (1999) 293–301 297

Society is prepared to pay large sums for low- In vegetated areas only 5–15% of the rainwater ered noise levels. Technical solutions to decrease runs off the ground, with the rest evaporating or noise include, for example, 3–5-m high walls at a infiltrating the ground. In vegetation-free cities cost of at least 5000 SEK (8 SEK:US$1) per m about 60% of the water is instead led off (Kommunfo¨rbundet, 1998). A wall like this de- through storm water drains (Bernatzky, 1983). creases the noise by 10–15 dB(A) immediately This will of course affect both the local climate behind it. However, the urban visual and the levels. Valuation of this would be destroyed if such walls were built every- service depends on the local situation. Cities with where. Another example of a technical solution is a high of flooding will benefit more from insulated windows in houses, but they are only green areas that take up water than do other effective for indoors. cities. In Stockholm, about 20% of the population is The in Stockholm is supplied by exposed to noise levels of over 55 dB(A) outside lake water. Therefore, the ground water levels in their homes, the maximum recommended level by the city are not heavily affected. Stockholm could the Swedish Environmental Protection Agency. however benefit from improved rainwater Some 630 km of streets have average roadside drainage through soft ground since the building noise levels of 60 dB(A) or more (Miljo¨fo¨rvaltnin- and maintenance of the storm water drainage gen, 1995). Increasing the areas with soft ground system involve large costs. Using the ecosystem and vegetation may decrease these noise levels. service could lower the cost. Vegetation may also contribute by shielding the visual intrusion of traffic and thus making it less 3.5. Sewage treatment disturbing: Evergreen trees are preferred in this case. Stockholm sewage treatment plants annually treat more than 150 million m3 of sewage (Stock- 3.4. Rainwater drainage holm Vatten, 1998). Taking care of sewage costs cities large amounts of , and the nutrients The built-up infrastructure, with concrete and that are still released contribute to tarmac covering the ground, results in alterations of the surrounding water ecosystems. of water flow compared to an equivalent rural In many cities, large scale experiments are tak- catchment. A higher proportion of rainfall be- ing place where natural systems, mainly wetlands, comes surface-water run-off which results in in- are being used to treat sewage water. The wetland creased peak flood discharges and degraded water plants and animals can assimilate large amounts quality through the pick-up of e.g. urban street of the nutrients and slow down the flow of the pollutants (Haughton and Hunter, 1994). The im- sewage water, allowing particles to settle out on pervious surfaces and high extraction of water the bottom. cause the groundwater level of many cities to Up to 96% of the nitrogen and 97% of the decrease. phosphorous can be retained in wetlands, and so Vegetated areas contribute to solving this prob- far wetland restorations have largely been success- lem in several ways. The soft ground of vegetated ful, increasing and substantially low- areas allows water to seep through and the vege- ering costs of sewage treatment (Ewel, 1997). The tation takes up water and releases it into the air cost of nitrogen reduction through wetland through evapotranspiration. restoration has been calculated to 20–60 SEK Even if the built city surface primarily seals the while the cost in a sewage treatment is ground from rainwater, it has been suggested that 33–350 SEK (Gren, 1995). Other benefits of wet- also creates some new, unintended , e.g. production and biodiversity, pathways for recharge. These include leaking wa- have not been included in these figures. ter mains, sewers, septic tanks, and soakways Stockholm has very few natural wetlands avail- (Lerner, 1990). able for sewage treatment, but it is possible to 298 P. Bolund, S. Hunhammar / Ecological Economics 29 (1999) 293–301 construct more wetlands for cleaning sewage wa- spaces can increase the physical and psychological ter. If all converted wetlands of the Stockholm well-being of urban citizens. catchment were restored, the cost of halving the The scientific values of ecosystems are also nitrogen load to the archipelago could be lowered included in this group, e.g. providing information by 20% (Gren, 1995). services. The urban ecosystems can function as indicators of the state of the urban environment. 3.6. Recreational and cultural 6alues Lichens, for example, cannot grow in areas with polluted air, and can thus be used to indicate the A city is a stressful environment for its citizens. air quality (Miller, 1994). The overall speed and number of impressions The citizens of Stockholm highly value their cause hectic lifestyles with little room for rest and green spaces: more than 90% visit parks at least contemplation. once during the year, 45% do so every week, and The recreational aspects of all urban ecosys- 17% more than three times a week (Stadbyggnad- tems, with possibilities to play and rest, are per- skontoret, 1994). In a stated preference study, haps the highest valued in cities. performed in Stockholm and a few other Swedish All ecosystems also provide aesthetic and cultural cities, people were willing to pay 360–530 SEK/ values to the city and lend structure to the land- month to live near a park, they were prepared to scape. Botkin and Beveridge (1997) argue that pay 370–540 SEK/month to live close to a larger ‘‘Vegetation is essential to achieving the quality of urban forest and 330–570 SEK/month to live life that creates a great city and that makes it close to water areas (Transek, 1993). possible for people to live a reasonable life within an urban environment’’. According to the Swedish Nils Lundgren, a good urban 4. Synthesis environment is an important argument for regions when trying to attract a highly qualified - In the previous section, the ecosystem services force (N. Lundgren, Nordbanken, personal were listed individually. It is however obvious that communication). each ecosystem generates a number of different The appearance of , e.g. birds and fish, services simultaneously. This is shown in a matrix should also be accounted for in recreational val- (Table 1) where we can see that all ecosystems ues. In Stockholm, a central stream of water contribute to climate regulation as well as provid- provides excellent opportunities for fish to spawn ing recreational and cultural values. Wetland also and the area is one of the best places to fish in the seems to be a valuable ecosystem type since it entire country. Approximately 30 different species contributes to all services. This corresponds to the are found here (Stadbyggnadskontoret, 1995). study by Costanza et al. (1997) where wetlands Green spaces are also psychologically very im- were ranked as the most valuable terrestrial portant. One example is a study on the response ecosystem per ha. of persons put under stress in different environ- If the aim is to assess the total value of ecosys- ments (Ulrich et al., 1991). This study showed tems in urban areas, it is important to add the that when subjects of the experiment were ex- value of all cells in a matrix of this kind. The posed to natural environments the level of stress individual values might be small, but taken to- decreased rapidly, whereas during exposure to the gether the total value of urban ecosystems is urban environment the stress levels remained high potentially significant. It should also be remem- or even increased. Another study on recovery of bered that the services discussed in this paper are patients in a hospital showed that patients with only a subset of the existing services. rooms facing a park had 10% faster recovery and The purpose of this paper is to analyze the needed 50% less strong pain-relieving medication benefits received from ecosystems, but ecosystems compared to patients in rooms facing a building can also cause problems. The main reason for wall (Ulrich, 1984). These studies imply that green building houses, as well as cities, has been to P. Bolund, S. Hunhammar / Ecological Economics 29 (1999) 293–301 299

Table 1 Urban ecosystems generating local and direct services, relevant for Stockholm.

Street tree Lawns/parks Urban forest Cultivated land Wetland Stream Lakes/sea

Air filtering X X X X X Micro climate regula-XX X X XXX tion Noise reductionXX X X X Rainwater drainage X X XX Sewage treatment X Recreation/cultural XX XXXX X values

protect humans from nature. The ecosystems kept the other hand are reported to recycle organic in cities contribute to urban well-being but may waste efficiently and produce much of their own also create negative aspects. Some common city food (Yufang et al., 1994). However, it is not tree species, for example pine (Pinus spp.), oak evident that more self-sufficient urban areas are (Quercus spp.), and willow (Salix spp.), emit simultaneously more sustainable. volatile organic compounds that may contribute Urban ecosystems are threatened by the process to urban smog and ozone problems (Slanina, of increasing the density of buildings. Trees are 1997). Animals, such as birds at municipal solid sometimes lost at a faster rate than they are waste dumps or frogs in wetlands, could cause replanted. The American Association disturbing noise and the restoration of wetlands found in a survey quoted in Moll (1989) that New could cause problems such as increased mosquito York City had a loss of approximately hatching and bad odors. The parks could be 175 000 street trees, or 25% of its total tree stand, dangerous places during the dark hours. In a during 1977–1987. In Stockholm about 8% of the complete cost-benefit analysis of land use and green space was lost during the , 7% during urban ecosystems, such negative aspects should the and, the process still continues in the also be reviewed. (La¨nsstyrelsen, 1996). Urban ecosystems are also often of poorer quality than their rural equivalents. By studying 5. Land use an urban-to-rural gradient in New York City, a scientific team discovered that forests at the urban One important issue in the debate on sustain- end of the gradient exhibited reduced fungal and able cities is whether expansion should be directed microarthropod populations and poorer leaf litter at increasing or rather allowing quality than the more rural forests (McDonnell et . Sprawled cities can produce more al., 1997). services while occupying a larger For the preservation of fauna, the size and amount of land. Even if a number of problems nature of the urban green areas are also impor- are created by the urbanization process, e.g. dis- tant. An area with a variety of biotopes will have rupted nutrient cycles and concentration of pollu- a large number of ecological niches that can be tants, urbanization also creates opportunities. If occupied by many different species, and will thus people live in dense concentrations, environmen- increase biodiversity. To have a high diversity of tally benign solutions like and plants and species in the city requires that the district heating become feasible (Rees and Wack- connections between the ecosystems surrounding ernagel, 1996). European cities are often dense the city and the green spaces in the city are not and to a large extent dependent on ecosystem disrupted. The small city parks and urban forests services from the outside. Some Chinese cities on are often too small to sustain a varied flora and 300 P. Bolund, S. Hunhammar / Ecological Economics 29 (1999) 293–301 fauna in themselves. Through the migration of Hopefully, an increased awareness of the organisms from larger core areas outside the city, ecosystem services could contribute to a more the diversity in urban ecosystems can still be resource-efficient city structure and design. The maintained. For example, Italian cities have been urban ecosystems could then be fully appreciated shown to contain almost 50% of all species of the for their contribution to urban life and valued total Italian avifauna (Dinetti et al., 1996) and accordingly when the land is claimed for exploita- over 1000 different vascular plant species have tion. An understanding of the importance of been identified in central Stockholm ecosystem services could also mean that unex- (La¨nsstyrelsen, 1996). However, the and ploited urban areas can be maintained or even railroads and large built-up areas around cities expanded. As cities are expected to grow at a often cause major barrier effects to the migration rapid rate in the coming decades, it is important of many species, and can thus lower the stabiliz- that the ecosystem services in urban areas and the ing effect of outer core areas (Bolund, 1996). ecosystems that provide them are understood and Since land is so valuable in urban areas, a valued by city planners and political decision- combination of different land uses on the same makers. piece of land is probably needed in order to safeguard and improve the generation of ecosys- tem services. Different strategies can be used to Acknowledgements increase vegetation, e.g. trees in parking spaces or narrow lawns as lane-separators. Some creative This paper has been written within the HUSUS thinking is needed. ( and Urban Structures in Sustainable Cities) project which is funded by BFR (Swedish Council for Building Research). Two anonymous 6. Concluding discussion reviewers provided useful comments on an earlier draft. We have tried to identify, and whenever possi- ble also quantify and value, the ecosystem services generated in urban areas. For most general ecosystem services, the share generated by ecosys- References tems within the urban area is expected to be Alexandersson, H., Karlstro¨m, C., Larsson-McCann, S., 1991. limited compared to the total service. However, Temperature and in Sweden 1961–1990. Ref- even if the generation of the services can often be erence normals, Report No. 81, SMHI, Norrko¨ping, 87 pp. made at a distance from the city, there are reasons Bernatzky, A., 1983. The effects of trees on the urban climate. why part of the services should be produced lo- In: Trees in the . Academic Publishers, cally. It can be advantageous to generate ecosys- Berkhamster, pp. 59–76 Based on the first International Arbocultural Conference. tem services locally for pure efficiency reasons, Bolund, P., 1996. Ecological Problems Caused by Roads and but also on ethical and educational grounds. Railroads. Master Thesis 1996:20. Department of Systems It is also clear that urban ecosystem services Ecology, Stockholm University, 30 pp. contribute to the quality of urban life even if Botkin, D.B., Beveridge, C.E., 1997. Cities as environments. urban citizens are still dependent on global Urban Ecosystems 1, 3–19. Bramryd, T., Fransman, B., 1993. Stadens lungor- om ecosystem services for their survival. The quality luftkvaliteten och va¨xtligheten i va˚ra ta¨torter (The lungs of of life for urban citizens is improved by locally the city—on air quality and vegetation in our cities). generated services, e.g. air quality and noise levels Movium-SLU Stad och Land 116, Alnarp (quoted from that cannot be improved with the help of distant Svensson and Eliasson 1997; in Swedish). ecosystems. It should however be remembered Costanza, R., d’Arge, R., de Groot, R., Farber, S., Grasso, M., Hannon, B., Limburg, K., Naeem, S., O’Neill, R., that it is only the effects of these problems that Paruelo, J., Raskin, R., Sutton, P., van den Belt, M., 1997. are decreased, not the cause of the problem that is The value of the world’s ecosystem services and natural solved. It is necessary to work to both ends. . Nature 387 (15), 253–260. P. Bolund, S. Hunhammar / Ecological Economics 29 (1999) 293–301 301

Dinetti, M., Cignini, B., Fraissinet, M., Zapparoli, M., 1996. Naturva˚rdsverket, 1996. Va¨gtrafikbuller. Nordiska bera¨kn- Urban ornithological atlases in Italy. Acta Ornithologica ingsmodeller (Roadnoise. Nordic calculation models). Re- 31, 15–23. port 4653, Stockholm, 110 pp. (in Swedish). Eurostat, 1995. Europe’s Environment: Statistical Com- Odum, E.P., 1971. Fundamentals of Ecology. Saunders, pendium for the Dobris Assessment. Eurostat, EU, UN Philadelphia 574 pp. OECD and WHO, Luxembourg 454 pp. Rebele, F., 1994. and special features of urban Ewel, K.C., 1997. improvement by wetlands. In: ecosystems. Global Ecol. Lett. 4, 173–187. Daily, G.C. (Ed.), Services. Societal Dependence Rees, W., Wackernagel, M., 1996. Urban ecological foot- on Natural Ecosystems. Island Press, Washington, DC, pp. prints: why cities cannot be sustainable - and why they are 329–344. a key to . Environ. Impact Assessment Rev. 16, 223–248. Folke, C., Jansson, A˚ ., Larsson, J., Costanza, R., 1997. Slanina, S. (Ed.), 1997. Biosphere-atmosphere exchange of Ecosystem appropriation of cities. Ambio 26 (3), 167–172. pollutants and trace substances. In: Transport and Chemi- Givoni, B., 1991. Impact of planted areas on urban environ- cal Transformation of Pollutants in the Troposphere, vol. mental quality: a review. Atmos. Environ. 25B (3), 289– 4. Springer, Berlin 528 pp. 299. SOU, 1993. Handlingsplan mot buller (Actionplan on noise). Gren, I.M., 1995. Costs and benifits of restoring wetlands. SOU 65, 136 Stokholm (in Swedish). Two Swedish case studies. Ecol. Eng. 4, 153–162. Stadbyggnadskontoret, 1994. Miniskrift om Stockholms Haughton, G., Hunter, C., 1994. Sustainable Cities, Regional parker och naturomra˚den (Brochure on parks and nature Policy and Development. Jessica Kingsley, London 357 pp. areas in Stockholm). SBK94:6. Stockholm, 10 pp. (in Hough, M., 1989. City Form and Natural Process. Routledge, Swedish). London 280 pp. Stadbyggnadskontoret, 1995. Stockholms ekologiska Kommunfo¨rbundet, 1998. Sko¨nheten och oljudet (The Beauty ka¨nslighet (The ecological sensitivity of Stockholm). SBK and the noise). Handbok i trafikbuller skydd, Svenska 1995:1. Stockholm, 62 pp. (in Swedish). Kommunfo¨rbundet, Stokholm, 132 pp. (in Swedish). Stockholm Vatten, 1998. Stockholms avloppsrening (Sewage La¨nsstyrelsen, 1996. Miljo¨analys 1996 Stockholms la¨n (Envi- treatment in Stockholm), Stokholm, 4 pp. (in Swedish). ronmental Analysis 1996 Stockholm County). Stolt, E., 1982. Vegetationens fo¨rma˚ga att minska exposi- La¨nsstyrelsen i Stockholms La¨n. Stockholm, 236 pp. (in tionen fo¨r bilavgaser (The ability of vegetation in decreas- Swedish). ing exposure to car fumes). Go¨teborgs Universitet pa˚ Lerner, D., 1990. in urban areas. In: uppdrag av Go¨teborgs Ha¨lsova˚rdsavdelning (quoted from Massing, H., Packman, J., Zuidema, F.C. (Eds.), Hydro- Svensson and Eliasson 1997, in Swedish). logical Processes and Water Management in Urban Areas, Svensson, M. and Eliasson I., 1997. Gro¨nstrukturens betydelse pp. 59–65 IAHS publ. no. 198, 1990. fo¨r stadens ventilation (The importance of green areas for McDonnell, M.J., Picket, S.T.A., Groffman, P., Bohlen, P., the ventilation of the city). Naturva˚rdsverkets rapport Pouyat, R.V., Zipperer, W.C., Parmelee, R.W., Carreiro, 4779, Stockholm (in Swedish). M.M., Medley, K., 1997. Ecosystem processes along an Tolly, J., 1988. Tra¨d och trafikfo¨roreningar samt Bil. Biolo- urban-to-rural gradient. Urban Ecosystems 1, 21–36. giskt filter fo¨r E4 pa˚ Hisingen (Trees and transport pollu- tion and the car). Go¨teborgs Stadsbyggnadskontor, McPherson, E.G., Nowak, D., Heisler, G., Grimmond, S., Hisingen, 15 pp. (quoted from Svensson and Eliasson, Souch, C., Grant, R., Rowntree, R., 1997. Quantifying 1997, in Swedish). urban forest structure, function and value: the Chicago Transek, 1993. Va¨rdering av miljo¨faktorer (Valuation of envi- Urban Forest Climate Project. Urban Ecosystems 1, 49– ronmental aspects). Transek, Solna (in Swedish) 115 pp. 61. Ulrich, R., 1984. View through a window may influence Miljo¨fo¨rvaltningen, 1995. Miljo¨ 2000 Miljo¨program fo¨r Stock- recovery from surgery. Science 224, 420–421. holm (Environment 2000 Environmental program for Ulrich, R.S., Simons, R.F., Losito, B.D., Fiorito, E., Miles, Stockholm). Fo¨rslag Feb. 1995, Stockholm. 93 pp. (in M.A., Zelson, M., 1991. Stress recovery during exposure to Swedish). natural and urban environments. J. Environ. Psychol. 11, Miller, G., 1994. Living in the Environment, 8th ed. 201–230. Wadsworth, Belmont, CA 701 pp. UN, 1997. Urban and Rural Areas 1996. UN, New York Moll, G., 1989. In search of an ecological urban landscape. In: publications (ST/ESA/SER.a/166), Sales Moll, G., Ebenreck, S. (Eds.), Shading Our Cities. Island No. E97.XIII.3, 1997. Press, Washington, DC 329 pp. Yufang, S., Tieheng, S., Ping, G., Zhijun, C., 1994. Resource- Moll, G., Petit, J., 1994. The urban ecosystem: putting nature ful ecological treatment of wastewater in urban ecosystems. back in the picture. Urban Forests Oct/Nov, 8–15. J. Environ. Sci. 6 (4), 487–495.

.