Urban Green Space Indicator

Usability and Accessibility of UGS in the City of Zurich and Uster

Introduction

Green public spaces in particular can contribute massively to a better quality of life (UN Habitat, 2013). They provide room for social and cultural interaction (UN Habitat, 2018) and are physically, socially and mentally beneficial to people's health (Peschardt et al., 2012). Several studies have come to the conclusion that especially accessible green spaces can reduce stress and generally lead to less health complaints (Stigsdotter et al., 2010; van den Berg et al., 2010). The influence of such green spaces is expected to be greater in urban environments, where people are usually confronted with poor air quality, water pollution or crowding (Nutsford et al., 2013). 

Such urban green spaces can function as a place for leisure, recreational activities, social cohesion, or cultural exchange (Belmeziti et al., 2018). Based on these findings (more elaborated in chapter “Conceptualization of Green Space”), we came up with an indicator to categorize urban green spaces into two main classes: leisure and recreation. Furthermore, we also took accessibility into account because we were interested in the walking time people had to put up with to reach a green space. Here, we further distinguished between mobility-impaired (e.g. wheelchair users) and not mobility-impaired people to make sure to leave no one behind. Mobility impairments have influence on the walkability, as natural roads or stairs are not accessible to people in a wheelchair.  

The indicator is calculated on a local level. Two cities were selected that lie in the canton of Zurich: Zurich City and Uster. Here, the goal was to compare two different urban areas. Zurich City represents a core city area, whereas Uster is located in the agglomeration of Zurich City (Statistik Stadt Zürich, 2017).  Zurich is the largest city of Switzerland and home to over 400'000 people. With its unique location on Lake Zurich, numerous recreational areas and the Alps on the horizon, Zurich City offers a wide range of attractions. In terms of quality of life, the city repeatedly achieves top marks in international studies (Stadt Zürich 2021). Uster is located 14 kilometers east of Zurich on Lake Greifensee. The municipality has over 35'000 inhabitants and is one of the 20 largest cities in Switzerland (Stadt Uster 2018b). The Greifensee nature reserve is the largest in the canton of Zurich (Stadt Uster 2018a).  

The goal of our indicator was to assess the accessibility of leisure and recreational green spaces in a certain urban area without leaving mobility-impaired people behind. On the DPSIR scale of the European Environmental Agency (EEA), the indicator would best fit the aspect of impact while efficiency would be the most suitable on the ABCDE taxonomy (also by EEA). The indicator is part of the 11th UN’s Sustainable Development Goal (SDG), especially target 7 which aims to provide universal access to safe and inclusive green and public spaces, particularly for women and children, older generations and persons with disabilities. 

Eventually, the following question should be answered in this report: 

How many inhabitants have access to urban green spaces used for leisure and recreation that are available for mobility-impaired and not mobility-impaired people in the City of Zurich and Uster?

For the detailed conceptualization of our indicator, various criteria had to be considered. In the following sections these criteria are described, justified and further exemplified at the case of our indicator.


Conceptualization

Green Space

There are different definitions of urban green spaces. They can include a variety of surfaces such as alluvial zones, street trees, blue spaces like ponds, lakes and rivers, private gardens or beaches. It becomes clear, that the different definitions are very context specific (WHO Regional Office for Europe (2016).  

The WHO defines urban green space as an urban are that is covered by vegetation of any kind. Following this definition, an urban green space includes smaller green space features (e.g. street trees), larger green spaces for social and recreational functions (e.g. parks, green belts and playgrounds) and also green spaces that are not publicly available (e.g. facades, green roofs) (WHO Regional Office for Europe 2017).  

An urban green space can function as a place for leisure, recreational activities, social cohesion, or cultural exchange (Belmeziti et al., 2018). While parks mainly provide some contact with nature inside a municipal area and can be used for activities and events (Peters, 2010; UN Habitat, 2019), green spaces characterized by wilderness are rather seen as places to escape from the pressures of the everyday life (Schuster et al., 2004). Based on these findings, we came up with an indicator which categorizes urban green spaces into two main classes: leisure and recreation. We defined leisure green spaces as characterized by infrastructure like barbecue areas, toilets, playgrounds, or benches which are in favor of meeting family and friends. Sufficient benches, the existence of playgrounds and good facilities were mentioned multiple times by visitors of green spaces in a study about the preference of urban green space characteristics in Portugal (Madureira et al., 2018). Hence, we integrated ‘social cohesion’ and ‘cultural exchange’ listed by Belmeziti et al. (2018) into the category of leisure. On the other hand, recreational green spaces were defined as places of wilderness, pure nature and tranquility where infrastructure is purposely absent. Tranquil environments are dominated by natural sounds and natural features such as vegetation, birds or hydrology. Conversely, anthropogenic sounds and features are at low levels (Marafa et al., 2018).  

The goal was to cover all possible leisure and recreational green spaces, that are publicly accessible, for example to water bodies, forests, public parks, cemeteries, floodplain areas and protected areas which include a variety of biodiversity-rich areas such as peatlands and migratory bird reserves.  We define green spaces within the border of Zurich City and Uster as urban green spaces.  

Noise Pollution

The United Nations name providing a calming environment as one important aspect of public spaces (UN Habitat, 2019). It is therefore important to consider noise levels when working with urban green spaces as well. 

Environmental noise is perceived as the third largest environmental burden after air pollution and second-hand smoke (Mohareb & Maassarani, 2019). Lower noise level generally allows more relaxation (Gozalo et al., 2018). Most of the environmental noise in the EU comes from traffic, especially road transportation (EEA, 2020). This is also the case for Switzerland, where 1.1 million people were influenced by road traffic noise during daytime in 2015 (BAFU, 2018). In addition, rail and aircraft noises are also present, though with less noise and frequency than road noise. (EEA, 2020).  

While most people just claim to be annoyed by noise exposure (Ali & Tamura, 2003; Mirzaei et al., 2012), noise abundance can also have severe health impacts, such as ischemic heart disease, high blood pressure or tinnitus which can further induce anxiety, distress or depression (WHO & JRC, 2011). It was also found that acute noise pollution can affect the long-term memory of children and their performance in solving difficult tasks (Evans et al., 1995).  

The perception of traffic noise is also influenced by the scenery as people discern traffic noise as less disturbing when they see natural features instead of urban scenes and buildings (Watts et al., 1999). Thus, green spaces can have a positive effect on noise mitigation. Still, noise level should be considered when classifying urban green spaces.  

The Health Council of the Netherlands (1994) published a detailed study on the effect of noise levels where the threshold of annoying road noise was set at 42 dB. Moudon (2009) and Cohen et al. (2014) rate noise pollution under 62 dB as normally acceptable regarding health concerns. Noise of 75 dB or higher are assessed to be dangerous for the physical and psychological wellbeing (UN Habitat, 2013; Cohen et al., 2014). 

Accessibility

The existence of green spaces alone is not enough to contribute to the physical, social and mental well-being of humans, as poorly accessible green spaces are visited less frequently. The existence of green spaces and their good accessibility are equally important (James et al., 2015). An effective delivery of services by green spaces is only possible, if an easy access by different modes of transport is possible (Giuliani et al., 2021). The goal of accessibility indices is the measurement of relative remoteness or proximity to different points of interest (POI), in our case urban green spaces. To get more accurate accessibility assessments, street network approaches are preferred over simple Euclidean distance methods, as these tend to overestimate general accessibility (Le Texler et al., 2018).  

In our project we want to measure the accessibility of urban green spaces by foot. Walkability is therefore a measure of the effectiveness of community design promoting movement by walking. The World Health Organization (WHO) and other institutions support the improvement of walkability within communities to reduce health problems and increase fitness and sustainability (Rattan et al., 2012).  


Implementation

For the implementation of the project we worked with both ArcGIS and QGIS. While we pre-processed, manipulated and combined the different data in ArcGIS, we used the QNEAT3 tool in QGIS, which allowed us to calculate the accessibility of the greenspaces. Figure 1 shows a first overview of the whole workflow. Please refer to the link below for a detailed explanation of the steps and the respective ArcGIS models that were created in the process. This allows for the reproducibility of our work.

Figure 1: Basic workflow of the steps included in the project.


Results

Not Mobility-Impaired Users

Leisure

Map 1: Leisure green spaces (in green) for the not mobility-impaired group and the three walking distances in blueish color for Zurich and Uster. 

The indicator for the not mobility-impaired leisure target group shows the highest accessibility for both cities: Zurich and Uster. Most people (389’866) living in Zurich have access to a leisure green space (LGS) within 10 minutes. 59’961 people have to walk for 10 to 20 minutes to enjoy a LGS. Only 40 people have access within 20 to 30 minutes. For Uster the pattern is quite similar. Most of the inhabitants (38’225 people) can reach a LGS in less than 10 minutes, 439 people within 10 and 20 minutes, and no one has to walk longer than 20 minutes to reach a LGS which is suitable for leisure activities. Overall, we have a high accessibility and a lot of leisure green spaces.   

Recreation

Map 2: Recreation green spaces (in green) for the not mobility-impaired group and the three walking distances in blueish color for Zurich and Uster. 

The not mobility-impaired recreation target group shows a smaller accessibility for both regions. For Zurich, 172’736 people have access to a recreational green space (RGS) within less than 10 minutes. 174’238 people can reach a RGS in between 10 and 20 minutes. 97’105 people must walk 20 to 30 minutes to be in a RGS. For Uster 31’048 people have access to a RGS within 10 minutes. 5’448 people and 2’139 people respectively have access between 10 – 20 minutes and 20 – 30 minutes to enjoy a recreational green space

Mobility-Impaired Users

Leisure

Map 3: Leisure green spaces (in green) for the mobility-impaired group and the three walking distances in blueish color for Zurich and Uster. 

In Zurich and in Uster there are many urban green spaces for the mobility-impaired group which uses green spaces for social or leisure activities. For Zurich 314’225 people have access to a leisure green space (LGS) within less than 10 minutes. 121’182 people can reach such a LGS between 10 and 20 minutes, whereas 5’766 people must walk 20 to 30 minutes. Uster, located in the agglomeration of Zurich City, shows a similar accessibility. 28’357 inhabitants are within a LGS in 10 minutes. 9’735 people can reach a LGS within 10 and 20 minutes. Only 31 people have to walk between 20 – 30 minutes in order to access a LGS. Overall, we have a high accessibility and a lot of leisure green spaces.   

Recreation

Map 4: Recreation green spaces (in green) for the mobility-impaired group and the three walking distances in blueish color for Zurich and Uster. 

The map for the mobility-impaired recreation target group shows the smallest accessibility pattern. For Zurich, 54’598 people have access to a recreational green space (RGS) within less than 10 minutes. 119’057 people can reach a RGS between 10 and 20 minutes. 122’242 people have to walk 20 to 30 minutes to be in a RGS. Most people have access to a RGS between 10 and 30 minutes in Zurich. In Uster, most people can walk to a RGS within 20 minutes (12’780 people for 0 – 10 min, 15’536 people for 10 – 20 min). A smaller fraction of 6’101 people must walk 20 – 30 minutes to be in a recreational green space

Table 1: Numbers of inhabitants with access to leisure and recreational green spaces, for mobility-impaired and not impaired people 

Looking at Table 1, one sees the results of the number of inhabitants lying in the isochrones of the urban green spaces. The not mobility-impaired leisure group shows the highest total accessibility of all four target groups, closely followed by the not-mobility impaired recreation and the mobility-impaired leisure group. The mobility-impaired recreation group shows the smallest accessibility. 

Map 5: Comparison of recreation (left) and leisure (right) spaces for not mobility-impaired users.

If we look at the sum of walking distances from 0 to 30 minutes, we see no differences between not mobility-impaired recreation and not mobility-impaired leisure group for Zurich and Uster in terms of inhabitants. Both groups (recreation and leisure) have nearly the same total numbers of inhabitants (see Table 1). The only difference is the distribution in the three walking distance categories. For the leisure group, most people have access within 10 minutes (applies to Zurich and Uster) as seen in Map 5. The recreation group shows a shift towards the 10 – 30 minutes category, especially seen in Zurich.  

Map 6: Comparison of recreation (left) and leisure (right) spaces for mobility-impaired users.

When comparing the mobility-impaired recreation group with the mobility-impaired leisure group a different pattern emerges. The mobility-impaired recreation group has lower total population number than the leisure group for both cities (see Table 1). Again, there are differences in the three walking distance categories, with the mobility-impaired leisure group reaching most residents within 10 minutes for both cities and the mobility-impaired recreation showing a shift towards the 10 to 30 minute distance category (see Map 6).  Between the two accessibility modes (either mobility or not mobility-impaired) differences can be stated too. In general, we observe a better accessibility for not mobility-impaired people for Zurich and for Uster (see Table 1). The difference between the accessibility modes is much larger in the recreational function. Between the two accessibility modes (either mobility or not mobility-impaired) differences can be stated too. In general, we observe a better accessibility for not mobility-impaired people for Zurich and for Uster (see Table 1). The difference between the accessibility modes is much larger in the recreational function.  Between the two accessibility modes (either mobility or not mobility-impaired) differences can be stated too. In general, we observe a better accessibility for not mobility-impaired people for Zurich and for Uster (see Table 1). The difference between the accessibility modes is much larger in the recreational function.  

Between the two accessibility modes (either mobility or not mobility-impaired) differences can be stated too. In general, we observe a better accessibility for not mobility-impaired people for Zurich and for Uster (see Table 1). The difference between the accessibility modes is much larger in the recreational function.  


Discussion

Interpretation of the accessibility results 

For both investigated municipalities, the total number of not impaired people with access to green spaces shows very small differences between the two green space types (Table 1). Thus, people from Zurich and from Uster have similar access to leisure and recreational green spaces. But it is clearly observable that green spaces for leisure activities are generally less far away than such for recreational purposes as most people reach leisure green spaces within a 10 minute walk for both municipalities. This is mostly due to our green space classification, which requires low noise levels for recreational spaces. Since city centers usually show quite significant noise pollution, people must walk longer to reach such tranquil areas. On the other side, leisure green spaces such as parks are a prominent features of city centers. They do not have to be tranquil, since people meet there to interact anyway. This interaction requires more infrastructure and therefore fits our definition of leisure. It seems reasonable that leisure green spaces are closer for most people, since they should allow people to feel a touch of nature, without having to travel to rural areas. 

For mobility-impaired people, however, leisure green spaces are significantly easier accessible than recreational areas. This seems to be due to the amount of recreational areas, which is significantly lower than the one of leisure green spaces. As shown before, walking distances are,also shorter for leisure green spaces than recreational ones when it comes to not impaired people. As mobility-impaired people are not able to make use of stairs, cartways or unpaved roads and they generally move at a slower speed, these disparities further increase. As a result, many people have to travel for more than 20 minutes and apparently in some cases even more than 30 minutes to reach a recreational green space. The difference between the two green space types is much smaller for Uster, which makes sense, since Uster is less spacious than the city of Zurich. 

In both municipalities, mobility-impaired people have less access to both green space types. Again, this is a consequence of the respective street network used for the walkability algorithm. As mobility-impaired people cannot access certain streets and their travelling speed is slower than the one for not impaired people, there is a larger group of inhabitants, who are not able to reach a recreational area within 30 minutes. Here, Uster is more suitable since it offers in terms of its size more recreational green spaces than Zurich and is generally smaller, resulting in shorter travelling times. Although, due to its population density, almost six times more not impaired people can reach a recreational space in Zurich. For mobility-impaired people, this factor decreases to 4 times, due to the previously stated reasons. 

For leisure green spaces, however, Zurich provides a lot of opportunities for mobility-impaired people, as many parks and other green areas with toilets, benches, firepits, playground or other infrastructure used for leisure activities, are available in the city. Moreover, the numbers of inhabitants who can reach leisure green spaces decreases with walking time. So, for a travelling time of 10 minutes, almost the same amount of mobility-impaired people can reach a leisure green space as this is the case for not impaired people. This once again shows the immense density of Zurich's leisure green spaces.  

Strengths of the Indicator 

A strong pro of our indicator is that we have calculated isochrones instead of a Euclidean distance, which show the real walking distance based on a street network. Also, the amount of considered criteria seems quite versatile to classify the green spaces. Thanks to detailed features provided by OSM, we could directly examine the social function of an UGS and particularly categorize leisure green spaces more specifically. For the noise levels, voluminous datasets with a good resolution were available, which enabled a detailed noise level classification. 

Limitations of the Indicator 

The indicator has of course some limitations. First of all, we worked with data that we did not collect by ourselves. Swisstopo is quite valid, but for example the OSM data are not as trustworthy as Swisstopo. Possibilities for a further study could be, that the OSM locations are checked, where one could visit the playgrounds, toilets, waste basket etc. It gets clear, that this is very time consuming and not feasible to check individual benches even in a municipality. To solve this, random POI could be checked to make an accuracy assessment of OSM datasets. Very high resolution satellite images could be also used. Our POI are rather small objects and could therefore be covered by larger objects like trees. Regarding OSM data, our categorization of the green spaces only took the availability of certain objects into account, while ignoring the size of the green spaces. Thus, a large park with only one fire pit is still categorized as leisure, if noise pollution is not too high. 

For the mobility-impaired target group, only hard types of pavements were considered and obstacles like stairs were excluded. The slope was not included in the calculation, but it is an important parameter that can be a limiting factor for mobility-impaired people. For a further study we propose to include the slope as an additional parameter. Moreover, the travelling speeds were set to a default value, therefore not representing all people as some may have motorized wheelchairs, while others get pushed by another person. There are also different kinds of mobility impairments, which can result in different travel speeds. 

Generally, we only looked at walking, which does not tell the full story about accessibility. Especially in cities, bicycles are a prominent way of travelling. Hence, people who seek recreation in tranquil spaces might be motivated to go by bike, which reduces travelling time significantly. Thus, accessibility of green spaces becomes even better, mainly favoring access to recreational areas, which are not so frequent in Zurich.  

Regarding the green spaces, there was no minimum value chosen. We accepted all green spaces regardless their size. Maybe it would be useful to set a threshold, so smaller green spaces get removed. Plus, one could normalize the green space size per capita, since the capability of an area is limited and it does not make sense, if many people have access to a small park, which gets quickly overcrowded. This is particularly important for recreational green spaces. If too many people are present, it loses its character of tranquility and cannot provide a recreational function anymore.  

Due to the intersection between the UGS and the street network, only green spaces with access to streets were selected and therefore used for the walkability assessment.  This is a possible limiting factor, as access to certain green spaces were not taken into account. People living near a forest patch would only have access to it, if a street directly leads into the forest. 

Further important to note are edge effects. As we calculated the walkability within the boundaries of Zurich and Uster, green spaces in the neighboring communities are left out of the calculation. In reality, people living near the municipal boundaries also have the option to seek recreation or leisure in green spaces in surrounding regions. This can especially lead to errors in Zurich, where urban area continues outside of the city border.   

As the noise level classification is based on sharp thresholds, a part of a forest with a noise pollution of 43 dB is already seen as not recreational anymore, although such a small difference probably would not be noticed by humans. Still, some boundaries had to be set and since noise was not the only criteria used for the green space categorization, no huge errors should have occurred here. 

Lastly, we noticed that the total population data in Table 1 exceed the inhabitants of Zurich and Uster. We expect an uncertainty in the STATPOP dataset, which is partly due to the resolution as people were counted per hectare. This led to a slight overestimation of population, which could have been further increased in the following calculation steps. Here, the maps shown in the results section, can visually prove that more or less the whole population has access to green spaces without having to walk more than 30 minutes. This is especially the case for not impaired people seeking for leisure green spaces. A better resolution would solve this problem. As our indicator is more about the comparison of two cities, this error isnot that decisive. 


Conclusion

The urban green space indicator showed the usability and the accessibility of UGS for Zurich and Uster based on four different target groups. The four different target groups are part of the “leaving no one behind” principle, which we tried to include in our study. We presented four different maps showing the green spaces and the isochrones for 0-10, 10-20 and 20-30 minutes walking distance. Our indicator would recommend Uster for mobility-impaired people, who especially prefer recreational green spaces and Zurich for the ones who like to socialize in leisure green spaces. For not impaired people, both cities are well-suited for leisure as almost the whole population reaches a leisure green space within a 30 minute walk at most. All in all, the smaller size of Uster also leads to generally shorter travelling times, which is particularly advantageous for recreational spaces. On the other hand, one has to forgo other characteristics the big city of Zurich offers. 

Even though we have highlighted the strengths of the indicator, several limitations have been pointed out and solutions have been proposed. Further research is needed to implement and test the proposed solutions. 


References

Ali, S. A., & Tamura, A. (2003). Road traffic noise levels, restrictions and annoyance in Greater Cairo, Egypt. Applied Acoustics64(8), 815–823.  

Belmeziti, A., Cherqui, F., & Kaufmann, B. (2018). Improving the multi-functionality of urban green spaces: Relations between components of green spaces and urban services. Sustainable cities and society43, 1-10. 

European Environment Agency (2020). Environmental noise in Europe - 2020. In European Environment Agency (Issue 22/2019). 

Evans, G. W., Hygge, T., & Bullinger, M. (1995). Chronic Noise and Psychological Stress.  Psychological Science6(6), 333–338.

Gozalo, G. R., Morillas, J. M. B., González, D. M., & Moraga, P. A. (2018). Relationships among satisfaction, noise perception, and use of urban green spaces. Science of the total environment624, 438-450. 

Giuliani, G., Petri, E., Interwies, E., Vysna, V., Guigoz, Y., Ray, N., & Dickie, I. (2021). Modelling Accessibility to Urban Green Areas Using Open Earth Observations Data: A Novel Approach to Support the Urban SDG in Four European Cities. Remote Sens., 13, 1-26. 

Hatuka, T. (2016). Public space. Protest Cultures: A Companion17, 284–293. 

James, P., Banay, R. F., Hart, J. E., & Laden, F. (2015). A Review of the health benefits of greenness.Current Epidemiology Reports2(2), 131–142. 

Le Texier, M., Schiel, K. & Caruso, G. (2018). The provision of urban green space and its accessibility: Spatial data effects in Brussels. PLoS ONE, 13, 1-17. 

Madureira, H., Nunes, F., Oliveira, J. V., & Madureira, T. (2018). Preferences for Urban GreenSpace Characteristics: A Comparative Study in Three Portuguese Cities. Environments, 5, 1-13. 

Marafa, L. M., Tsang, F., Watts, G., & Yuan, X. (2018). Perceived tranquility in green urban open spaces. World Leisure Journal, 60, 1-15. 

Mohareb, N., & Maassarani, S. (2019). Assessment of street-level noise in three different urban settings in Tripoli. Urban Climate29, 100481.

Nutsford, D., Pearson, A. L., & Kingham, S. (2013). An ecological study investigating the association between access to urban green space and mental health. Public health127(11), 1005-1011. 

Peschardt, K. K., Schipperijn, J., & Stigsdotter, U. K. (2012). Use of small public urban green spaces (SPUGS). Urban forestry & urban greening11(3), 235-244. 

Peters, K. (2010). Being together in urban parks: Connecting public space, leisure, and diversity. Leisure Sciences32(5), 418-433. 

Raffler, C. (2018). QGIS Network Analysis Toolbox 3. <https://root676.github.io/> (Access 05.06.2021) 

Rattan, A., Campese, A., & Eden, C. (2012). Modeling Walkability, <https://www.esri.com/news/arcuser/0112/modeling-walkability.html> (Access 17.06.2021). 

Schuster, R., Tarrant, M., & Watson, A. (2004). The social values of wilderness. In In: Murdy, James, comp., ed. Proceedings of the 2003 Northeastern Recreation Research Symposium; 2003 April 6-8; Bolton Landing, NY. Gen. Tech. Rep. NE-317. Newtown Square, PA: US Department of Agriculture, Forest Service, Northeastern Research Station: 356-365. 

Stadt Uster (2018a): Der Greifensee, < https://www.uster.ch/geografiesee > (Access 14.06.2021). 

Stadt Uster (2018b): Geografie, < https://www.uster.ch/praesentationgeografie > (Access 14.06.2021). 

Stadt Zürich (2021): Über Zürich, < https://www.stadt-zuerich.ch/portal/de/index/portraet_der_stadt_zuerich.html > (Access 14.06.2021). 

Statistik Stadt Zürich (2017). AGGLOMERATION ZÜRICH, <https://www.stadt-zuerich.ch/content/dam/stzh/prd/Deutsch/Statistik/Publikationsdatenbank/jahrbuch/2017/pdf/JB_2017_kapitel_20.pdf> (Access 17.06.2021). 

Stigsdotter, U. K., Ekholm, O., Schipperijn, J., Toftager, M., Kamper-Jørgensen, F., & Randrup, T. B. (2010). Health promoting outdoor environments-Associations between green space, and health, health-related quality of life and stress based on a Danish national representative survey. Scandinavian journal of public health38(4), 411-417. 

UN Habitat (2013). Streets as public spaces and drivers of urban prosperity. Nairobi: United Nations. 

UN Habitat (2018). Public Spaces. Monitoring and reporting on the SDGs. Nairobi: United Nations. 

Van den Berg, A. E., Maas, J., Verheij, R. A., & Groenewegen, P. P. (2010). Green space as a buffer between stressful life events and health. Social science & medicine70(8), 1203-1210. 

WHO Regional Office for Europe (2011). Burden of disease from environmental noise: Practical guidance. Copenhagen: WHO Regional Office for Europe, <https://www.euro.who.int/en/health-topics/environment-and-health/noise/publications/2011/burden-of-disease-from-environmental-noise-practical-guidance>. 

WHO Regional Office for Europe (2016). Urban green spaces and health: a review of evidence. Copenhagen: WHO Regional Office for Europe, <http://www.euro.who.int/ en/health-topics/environment-and-health/urbanhealth/publications/2016/urban-green-spaces-andhealth-a-review-of-evidence-2016>. 

WHO Regional Office for Europe (2017): Urban green space interventions and health: a review of impacts and effectiveness. Copenhagen: WHO Regional Office for Europe <http://www.euro.who.int/__data/assets/ pdf_file/0010/ 337690/FULL-REPORT-for-LLP.pdf?ua=1>. 

Figure 1: Basic workflow of the steps included in the project.

Table 1: Numbers of inhabitants with access to leisure and recreational green spaces, for mobility-impaired and not impaired people