249Kolcsár, R.A. et al. Hungarian Geographical Bulletin 71 (2022) (3) 249–269.DOI: 10.15201/hungeobull.71.3.3 Hungarian Geographical Bulletin 71 2022 (3) 249–269. Introduction The term urban green space (UGS) is defined by World Health Organization as an urban area covered by any kind of vegetation (WHO 2017; WHO Regional Office for Europe 2017). UGSs are among the main contributors to hu- man well-being within a city (Neuvonen, M. et al. 2007; James, P. et al. 2009; Schipperijn, J. et al. 2010; Rossi, S.D. et al. 2015; Kothencz, G. et al. 2017; Kovács-Győri, A. et al. 2018; Csete, Á.K. et al. 2021). UGSs also contribute to the liveability of a settlement through a high va- riety of positive effects, which are collec- tively known as ecosystem services (MEA 2005; Makovníková, J. et al. 2019; CICES 2022). Via evapotranspiration and shading, vegetation impacts microclimates (MEA 2005; Takács, Á. et al. 2016; CICES 2022; TEEB 2022). The urban heat island effect, which is an increasingly sig- nificant problem in modern cities can also be mitigated by UGSs (Gál, T. et al. 2016; Henits, L. et al. 2016; Herbel, I. et al. 2016; Pongrácz, R. et al. 2016; Mucsi, L. et al. 2017). Other im- portant regulating functions of UGSs include surface-runoff mitigation, the enhancement of air quality, as well as wind speed and noise re- duction (MEA 2005; Csete, Á.K. and Gulyás, Á. Age-group-based evaluation of residents’ urban green space provision: Szeged, Hungary. A case study Ronald András KOLCSÁR1, Ákos K. CSETE2, Anna KOVÁCS-GYŐRI3 and Péter SZILASSI1 Abstract Analysis of urban green space (UGS) provision is becoming increasingly important from an urban-planning per- spective, as processes related to climate change tend to worsen the urban heat-island effect. In the present study, we aimed to map the UGS provision of Szeged, Hungary, using a GIS-based complex approach. Different age groups, especially the elderly, have different demands on the ecosystem services and infrastructure of UGSs. To provide an in-depth assessment of UGS provision for planners, we analysed the UGS availability and accessibility, using sub- block-level population data, which includes not only the total number of residents but also provides information about the age-group distribution for each building of the city. We delineated areas having different UGS provision levels (called provision zones) and assessed the age distribution of the residents living in each zone. We found that the residents within 2-min walking distance to public green spaces are older than expected by comparison to the age distribution of Szeged. In provision zones with abundant locally available UGSs (measured as UGS per capita within 50-m buffers), we found that the youngest (0–18 years) and oldest (≥ 61 years) inhabitants are overrepresented age groups, while the age group 19–40 has the lowest overall UGS provision within the city of Szeged. Our research, which has the potential to be adapted to other settlements, contributes to the identification of UGS-deficit areas in a city, thereby providing essential information for urban planners about where increases in UGS are most needed and helping to assess infrastructural enhancements that would be adequate for the locally most-dominant age groups. Keywords: availability, accessibility, provision, urban park, green infrastructure Received June 2022, accepted September 2022. 1 Department of Geoinformatics, Physical and Environmental Geography, University of Szeged. Egyetem u. 2–6. H-6722 Szeged, Hungary. Correspondent author’s e-mail: kolcsar@geo.u-szeged.hu 2 Institute of Ecology and Botany, Centre for Ecological Research, Alkotmány u. 2-4. H-2163 Vácrátót, Hungary. 3 IDA Lab, Paris-Lodron-Universität Salzburg, Jakob-Haringer-Str. 6. 5020 Salzburg, Austria. Kolcsár, R.A. et al. Hungarian Geographical Bulletin 71 (2022) (3) 249–269.250 2021; CICES 2022; TEEB 2022). Additionally, vegetation in private and community gardens may also contribute to food provision (MEA 2005; CICES 2022; TEEB 2022). In the form of cultural ecosystem services, UGSs help to create an aesthetic and calm environment ide- al for both physical recreation and spiritual rejuvenation (MEA 2005; Razak, M.A.W.A. et al. 2016; Ayala-Azcárraga, C. et al. 2019; Szilassi, P. et al. 2020; Cheng, Y. et al. 2021; CICES 2022; TEEB 2022). Considering all the advantages that UGSs provide, it is not unexpected that recently the issue of proper UGS provision has been receiving ever-growing attention within the field of urban planning (Zepp, H. et al. 2020). However, the definition and measurement of UGS provision and access are not uniform among researchers (Wolch, J.R. et al. 2014; Ekkel, E.D. and de Vries, S. 2017; Wüstemann, H. et al. 2017; Blaschke, T. and Kovács-Győri, A. 2020; Wu, L. and Kim, S.K. 2021). For exam- ple, Le Texier, M. et al. (2018) specified four levels of UGS provision and access: availabil- ity, fragmentation, public-private ownership, and accessibility. Other researchers identified five components of green space access, a con- cept similar to UGS provision: virtual access, viewing, utilizing and being in green space, active hands-on engagement, and ownership and/or management (Weldon, S. et al. 2007; Edwards, D. et al. 2009). One of the most well-systematized defi- nitions in connection with UGS provision comes from Biernacka, M. and Kronenberg, J. (2019). Their approach involves three lev- els of UGS provision: availability, accessibil- ity, and attractiveness. Among these three, UGS availability is the most fundamental, as it indicates whether green space exists within a given proximity. The most common method for the evaluation of UGS availability is the calculation of the total area of UGSs within a buffer zone; it is most commonly measured in hectares or square metres per capita (Fuller, R.A. and Gaston, K.J. 2009; Kabisch, N. and Haase, D. 2014; Kabisch, N. et al. 2016; Biernacka, M. and Kronenberg, J. 2019; Biernacka, M. et al. 2020; Kolcsár, R.A. et al. 2021). Csomós, G. et al. (2021), however, used different, more-complex methodologies to quantify UGS availability. Owing to various physical or psychological barriers, the second level of UGS provision, UGS accessibility, is not guaranteed to all or to part of the population for any particular existing (and, thus, available) UGS (Wright, W.H.E. et al. 2012; Park, K. 2017; Biernacka, M. and Kronenberg, J. 2019; Biernacka, M. et al. 2020; Kolcsár, R.A. et al. 2021). Physical accessibility means the lack of physical bar- riers (e.g., busy streets, fences or buildings) along the road that can potentially hinder or preclude access to a given green space (Biernacka, M. and Kronenberg, J. 2019). Buildings and road networks also can highly modify pedestrian travel times by making the route between the starting point and destination far longer than the straight-line distance between the two points (Kolcsár, R.A. et al. 2021). Psychological accessibility is defined by more-abstract barriers, such as the real or perceived relative safety of areas that some might choose to avoid, thereby lengthening their travel time to the UGS (Park, K. 2017). UGS accessibility is usually mapped either within buffer zones (Oh, K. and Jeong, S. 2007; Wright, W.H.E. et al. 2012; Braquinho, C. et al. 2015; Bahrini, F. et al. 2017; Koprowska, K. et al. 2018; Csomós, G. et al. 2020) or with more-accurate meth- ods (such as service areas) based on road networks (Shahid, R. et al. 2009; Le Texier, M. et al. 2018; Mora-Garcia, R.T. et al. 2018; Kolcsár, R.A. et al. 2021), and even more elaborate methods based on least cost path algorithms (e.g., using AccessMod and CostDistance) have been devised, such as ap- plied by Chênes, C. et al. (2021). Among the three levels of UGS provision, attractiveness is arguably the most difficult to objectively measure. Any available and freely accessi- ble green space needs to have certain quali- ties in order to attract residents to visit the area. UGS attractiveness is determined by various factors, which may involve cultural ecosystem services, aesthetic values, and infrastructure among others (Kronenberg, 251Kolcsár, R.A. et al. Hungarian Geographical Bulletin 71 (2022) (3) 249–269. J. 2015; Biernacka, M. and Kronenberg, J. 2019). The attractiveness of an UGS is shown to degrade by the increase of walking dis- tance (Morar, T. et al. 2014), while some UGSs are simply unattractive to visitors (Wright, W.H.E. et al. 2012; Biernacka, M. and Kronenberg, J. 2019). In some studies, the terms UGS quantity and UGS quality are considered to correspond to the concepts of availability and attractiveness, respectively (You, H. 2016; Kraemer, R. and Kabisch, N. 2021). Evaluation of availability and ac- cessibility of sectors other than UGSs (e.g., health care) is also possible with numer- ous occurrences in the scientific literature (Hare, T.S. and Barcus, H.R. 2007; Kwan, M.P. and Weber, J. 2008; McGrail, M.R. and Humphreys, J.S. 2009; Kraft, S. 2016; Uzzoli, A. et al. 2020). Van den Bosch, C.K. (2021) proposed the 3-30-300 rule, as a very similar concept of UGS provision. According to this rule, ideally, there should be at least three trees visible from one’s window, while at the same time one’s neighbourhood should have at least 30 percent UGS coverage (preferably trees), and one’s home should be no farther, than 300 m from the nearest urban park. Most authors agree that, depending on its size and function, a walking distance of 2–15 min to get to a public park is optimal for de- fining a catchment or area within which lo- cal residents can easily access public green spaces (Stanners, D. and Bourdeau, P. 1995; Barbosa, O. et al. 2007; Pafi, M. et al. 2016; Stessens, P. et al. 2017; Poleman, H. 2018; Kolcsár, R.A. et al. 2021). Although the grow- ing number of spatial data sources and GIS analysis techniques make it easier to deter- mine the number, and demographic charac- teristics (age, gender distribution, etc.) of the local inhabitants living within the different catchment areas of the public parks (Bok, J. and Kwon, Y. 2016; Wüstemann, H. et al. 2017; Poleman, H. 2018; Zepp, H. et al. 2020), to date little information is available on the age struc- tures of different UGS provision zones. The aim of the present study has been to investigate the age structure of areas with dif- ferent UGS provisions in a Hungarian study area, specifically as a case study of Szeged. We sought to evaluate the UGS provision at the building-plot scale with a method that in- corporates two of three UGS provision levels (availability and accessibility) identified by researchers (Biernacka, M. and Kronenberg, J. 2019; Kolcsár, R.A. et al. 2021). Given that the equality of UGS provision is an impor- tant component of environmental justice (Wen, C. et al. 2020), this method could potentially help urban planners to localize deficiency in the green infrastructure with the consideration of the age-related needs of local residents (Kemperman, A.D.A.M. and Timmermans, H.J.P. 2007; Artmann, M. et al. 2017; Levy-Storms, L. et al. 2018; Bozkurt, M. 2021; Yang, L. et al. 2021). Based on our over- all aims, we set out to answer the following research questions in this research: – How can we specify and delineate the low-, medium-, and high-level UGS provi- sion zones? – What is the age structure of local inhab- itants living in low-, medium-, and high- level UGS provision zones? Answering these questions is important from the perspective of urban planning, as it will help in the delineation of UGS-deficit areas within cities and, thus, where the im- provement of UGS provision is necessary. Analysing UGS provision by age group is also essential for both environmental justice and urban park design. Materials and methods Study area The study area of the present research, Szeged (Figure 1), is in the centre of the Southern Great Hungarian Plain (EU NUTS2 level statistical region of Hungary). By pop- ulation, Szeged is the third largest city in Hungary and the second largest in the Great Hungarian Plains (KSH 2021). The radio-concentric city has multiple ur- ban parks of various sizes and structures, the largest of which (~ 20 ha), Erzsébet Liget Kolcsár, R.A. et al. Hungarian Geographical Bulletin 71 (2022) (3) 249–269.252 (Elisabeth Park) is located on the western side of Újszeged district. Besides its urban parks, Szeged has a notable amount of forest areas as well. These are decisively floodplain forests along the bank of its two rivers, Tisza and Maros. Informal green spaces (especially in the housing estate areas) and private gar- dens also constitute a significant part of the green infrastructure of Szeged. Data used A detailed address-level population database was used both for local UGS availability (as per-capita estimations) and for demographic analyses of the UGS provision zones (Bel- ügyminisztérium 2019). To enable spatial evaluations and visualization, a polygon layer containing Szeged’s building plots with permanent residents was also created (see Figure 1). Layer contained information solely about the population with permanent residency in Szeged, thus, other forms of res- idency (e.g., renters, students living in dor- mitories etc.) were not included in this study. For local green space availability map- ping, we used a high-resolution Normalized Difference Vegetation Index (NDVI) layer (0.5 x 0.5 m raster size) derived from an ortho- photo (Lechner Tudásközpont, 2015). Public green space accessibility mapping required the usage of layers (e.g., road network and points of interest) from the open-source GIS database OpenStreetMap (Geofabrik, 2021). For both the local green space availability and public green space accessibility analyses, we incorporated the land-use and land-cover polygons of the 2018 version of the European Environmental Agencies Urban Atlas (Coopernicus, 2018). Figures of the present paper were produced using data provided by Copernicus (Urban Fig. 1. Study area within Szeged, Hungary, showing building-plot-scale population distribution and public green spaces selected for accessibility estimation. 253Kolcsár, R.A. et al. Hungarian Geographical Bulletin 71 (2022) (3) 249–269. Atlas), Geofabrik (OpenStreetMap), Lechner Tudásközpont (orthophoto of Szeged), the Hungarian Ministry of the Interior (address- level population data), and the 2018 buil- ding-code modification document of Szeged (building plots) (Lechner Tudásközpont 2015; Copernicus 2018; Fehér, É. 2018; Belügyminisztérium 2019; Geofabrik 2021). Population data pre-processing Building plots were chosen as the adminis- trative unit for our study as they provide a realistic picture of the spatial extent of living space of the urban population (particularly for detached houses). The polygon layer contain- ing building plots was digitized from maps in the 2018 building-code modification document for Szeged (Fehér, É. 2018). At our disposal was a building-polygon database that contained highly accurate, address-level population data (capita per building) based on the postal mail- ing addresses of all inhabitants of Szeged as of 1 January 2019 (Belügyminisztérium 2019). The building-polygon layer with population data was spatially joined to the polygon layer of the building plots. Generally, one building plot contains only one building, in the case of building-plot polygons containing two or more buildings, however, we used the sum of popu- lation values of all buildings within its area for the building-plot population. By joining the two layers, building plots with no permanent resi- dents were deleted from the layer. Age-group categories in the source data were merged into larger categories to reduce fragmentation for the current analyses, thereby creating four new age groups: minors (0–18 years), young adults (19–40 years), middle-aged adults (41–60 years), and the elderly (≥ 61 years) (Figure 2), with dif- ferently attributed UGS usage habits and de- sign requirements (Kemperman, A.D.A.M. and Timmermans, H.J.P. 2007; Artmann, M. et al. 2017; Levy-Storms, L. et al. 2018; Bozkurt, M. 2021; Yang, L. et al. 2021). The division of these four merged categories was restricted by the original, narrower age groups, which had very inconsistent ranges regarding the span of years (e.g., 0–2 years, 3 years, 7–10 years, 41–45 years, 63–70 years, etc.). According to our popula- tion database, there were ~144,000 permanent residents within the study area (Szeged) on January 1, 2019. Local green space availability analysis We defined local green space availability (the local availability of any type of green space) as the total area of UGSs within 50-m-radius buffer zones of each building plot divided by the building plot’s number of residents (square metres per capita). The size of the buffer zone was chosen according to J. Gehl’s thresholds (2010), which were based on con- sideration of human-scale distance. To delineate areas covered with vegetation within the city, we created a raster-based NDVI map from a high-resolution orthophoto (0.5 x 0.5-m raster size) provided by the University of Szeged Department of Climatology and Landscape Ecology. The image was acqu- ired during May–July 2015 by Lechner Tudásközpont (2015). The footage was a four- band (RGB–NIR) UltraCam X orthophoto that had an appropriate resolution for highly accu- rate NDVI maps (Figure 3). 0 was chosen as the pixel value that represents minimal vegetation. Although the 0.2 value defined by the United States Geological Survey (USGS) is the most commonly accepted lower threshold of veg- etation coverage (USGS 2018), Aquino, D. et al. Fig. 2. The total number of residents within investi- gated age groups of Szeged, Hungary. Kolcsár, R.A. et al. Hungarian Geographical Bulletin 71 (2022) (3) 249–269.254 (2018) argued that it is possible that areas with very low vegetation coverage to fall between NDVI values of 0 and 0.2. In the case of our orthophoto, the most realistic delineation of vegetation coverage was achieved by apply- ing 0 as the minimum value. The suspected reason for a significant amount of UGSs was being found between values 0 and 0.2, was the heat stress that the vegetation suffered at the time of the data collection. Because of their seasonality and meager recreational val- ue, plow fields were excluded from the NDVI map. To mask out these undesirable areas, we used the European Environmental Agencies Urban Atlas 2018 land-use and land-cover da- tabase’s annual-crops polygons (code 21000) (Copernicus 2018). After the selection procedure, 50-m buffer zones were generated around polygons enclos- ing building plots with permanent residents, and the sum of green pixels (NDVI ≥ 0) was calculated for each 50-m buffer zone (Figure 4). We calculated total vegetation coverages (as square meters) from the raster resolution and the sums of pixels within the buffer zones. Dividing the resulting values by the number of residents for each corresponding building plot, we determined how much local green space is available for an inhabitant of a given building plot within its 50-m buffer zone. We then delineated zones with low, me- dium, and high local green space availability (local green space availability zones) based on the previously calculated square metres per capita values. We chose 50 m2 per capita as the boundary value between low and me- dium categories based on the the recommen- dations of World Health Organization (2017) which is widely used in various publications (Morar, T. et al. 2014; Maryanti, M.R. et al. 2016). As the boundary value between me- dium and high categories, we used 500 m2 Fig. 3. Normalized Difference Vegetation Index (NDVI) layer generated from 2015 orthophoto of Szeged, Hungary. 255Kolcsár, R.A. et al. Hungarian Geographical Bulletin 71 (2022) (3) 249–269. per capita derived from the optimal mini- mum green space availability defined for use in Berlin (5,000 m2 per capita within 500-m buffer zones) by converting it to the 50 m buffer zones of the present study (Kabisch, N. et al. 2016). A score between values 1–3 was also assigned to the local green space availability categories (Table 1). Accessibility mapping of public green spaces For the accessibility analysis, we created a subgroup of UGSs called public green spac- es, which aims to include every urban park, public urban forest or any other functionally similar area. Additionally, we aimed to in- clude in this category informal green spaces Fig. 4. Principle of calculation of local green space availability, demonstrated via example building plot. NDVI = Normalized Difference Vegetation Index. Table 1. Category limits of local green space availability and public green space accessibility Indicator Low (score 1) Medium (score 2) High (score 3) Local green space availability, m2 per capita Public green space accessibility, min. < 50 > 10 50–500 2–10 > 500 < 2 Kolcsár, R.A. et al. Hungarian Geographical Bulletin 71 (2022) (3) 249–269.256 as well, which are known contributors to the urban quality of life (Rupprecht, D.C. et al. 2015). Thus, on the basis of Kolcsár, R.A. et al. (2021) we defined public green spaces as areas which meet all of the following conditions: 1. The area must be defined by Urban Atlas (Copernicus 2018) land use and land cover database as either “Green urban areas” (14100), “Forests” (31000) or “Land without current use” (13400). 2. The area must be a public space. 3. The area has functions or services impor- tant enough to the public that it’s reflected in OpenStreetMap’s Point of Interest (POI) layer (Geofabrik 2021). Because, by definition, public green spaces are open to everyone, accessibility analysis can be used to incorporate them in a city- scale UGS provision assessment. Private gar- dens on the other hand, by their nature are inaccessible to the public and, thus, were not included in the accessibility analysis. The first step of our selection process was to select polygons defined in Urban Atlas as Green urban areas, Forests or Land without current use. The latter category was included because of its potential to contain informal green spaces. As the next step, we overlapped the point of interest (POI) layer, which con- sists of volunteered geographic information (Jiang, S. et al. 2015; Zhang, X. et al. 2017) with the selected UGSs to find the areas with at least one function or service significant enough to appear in the publicly accessible and editable database. In most cases, POI-s within these USG polygons represented ob- jects connected to amenities (e.g., street fur- niture, drinking water, and toilets), recrea- tion (e.g., playgrounds) or aesthetic values (e.g., statues, fountains, and landmarks) all of which can be connected to the multi-func- tionality of the USG. Because of this, the POI layer proved to be an invaluable tool to iden- tify and delete UGS polygons with little to no functions. One UGS polygon was retained al- though it is not notably represented in the POI layer in the database. While none of the POI layer’s points intersected its geometry, a point was found within its direct proximity which was identified as a POI of this particular pub- lic green space. The remaining polygons were individually evaluated as to whether or not they are publicly accessible, and all private areas were excluded from further analysis. On the basis of the work of Stessens, P. et al. (2017) and Kolcsár, R.A. et al. (2021), we divided a selection of public green spaces into two separate groups by size: local public green spaces (0.3–10 ha) and district public green spaces (10–1,000 ha). We then assigned to each group a maximum walking distance to represent its catchment size: 200 m for lo- cal public green spaces and 800 m for district public green spaces. Following Kolcsár, R.A. et al.’s method- ology (2021), the OpenStreetMap road net- work was overlapped with the public green space polygons (Geofabrik 2021), and entry points were generated at the intersections of the road lines and the polygons. With the help of ArcGIS Pro’s Network Analyst tool (ESRI 2018), service areas (catchments) were generated around these entry points to in- dicate the maximum pedestrian travel time required to reach any given point within the city. Assuming a 5-km/h pedestrian velocity, catchments were converted into travel times: ~2 min in the case of local public green spac- es and ~10 min in the case of district public green spaces. Accordingly, catchments repre- senting 2 min walking distance were gener- ated around each of the entry points, while catchments representing 10 min walking dis- tance were generated only around the entry points of district public green spaces. By merging these catchments together (2-minute catchments and 10-minute catchments sepa- rately), the public green space accessibility zones were created. With the help of the Select Layer by Location tool of ArcGIS (using the centroids of each building-plot polygon with perma- nent residents), public green space accessi- bility zones were projected to the building plot layer. Each building plot was then cat- egorized based on their public green space accessibility. Building plots with a centroid within < 2-min public green space accessibil- 257Kolcsár, R.A. et al. Hungarian Geographical Bulletin 71 (2022) (3) 249–269. ity zones were evaluated as exhibiting high public green space accessibility, and those within 2- to 10-min public green space ac- cessibility zones were evaluated as exhibit- ing medium public green space accessibility. Any building plots outside the public green space accessibility zones (> 10-min) were de- fined as having low public green space ac- cessibility. A score between the values of 1–3 was also assigned to the public green space accessibility categories (see Table 1). Overall UGS provision assessment Using the local green space availability and the public green space accessibility scores (1–3), we implemented a combined scoring system to evaluate the overall UGS provision of each building plot (Table 2). From these scores (2–6), we created the overall UGS pro- vision categories. A three-level classification was created so the overall UGS provision results would be comparable with the local green space availability and public green space accessibility estimates. A five-level classification was also created to illustrate finer differences between provision zones. Based on Table 2, overall UGS provision zones (and the corresponding scores) can be interpreted as follows: Low UGS provision zones (score 2-3): 1. Areas with no or very little locally avail- able UGS (below 50 m2/capita) and no acces- sible public green space within 10 minutes of walking. 2. Areas with no or very little locally availa- ble UGS (below 50 m2/capita), but the nearest public green space can be accessed between 10 and 2 minutes by walk. 3. Areas with an acceptable amount of lo- cally available UGS (between 50 and 500 m2/ capita) but no accessible public green space within 10 minutes of walking. Medium UGS provision zones (score 4): 1. Areas with no or very little locally avail- able UGS (below 50 m2/capita) but the near- est public green space can be accessed within 2 minutes by walk. 2. Areas with an acceptable amount of lo- cally available UGS (between 50 and 500 m2/ capita) and an accessible public green space between 10 and 2 minutes by walk. 3. Areas with an abundance of locally available UGS (500 m2/capita or higher), but no accessible public green space within 10 minutes of walking. High UGS provision zones (score 5 and 6): 1. Areas with an acceptable amount of lo- cally available UGS (between 50 and 500 m2/ capita) and the nearest public green space can be accessed within 2 minutes by walk. 2. Areas with an abundance of locally available UGS (500 m2/capita or higher), and an accessible public green space between 10 and 2 minutes by walk. 3. Areas with an abundance of locally avail- able UGS (500 m2/capita or higher), and the nearest public green space can be accessed within 2 minutes by walk. Age-group distributions of UGS provision zones We performed age-group-based demograph- ic analyses of the data within the building- plot polygons. As a result of the overall UGS provision evaluations, the age-group distri- bution could be estimated for the three UGS provision zones (low, medium, and high) as Table 2. Numeric evaluation of overall urban green space provision Indicator Public green space accessibility scores Local green space availability scores Low (score 1) Medium (score 2) High (score 3) Low (score 1) very low (score 2) low (score 3) medium (score 4) Medium (score 2) low (score 3) medium ( score 4) high (score 5) High (score 3) medium (score 4) high (score 5) very high (score 6) Kolcsár, R.A. et al. Hungarian Geographical Bulletin 71 (2022) (3) 249–269.258 well as separately for the local green space availability and the public green space ac- cessibility zones. In addition to the actual population numbers (pa), we also calculated an estimated population number (pe) (Eq. 1), showing what would be the number of resi- dents within each age group if the age distri- bution of the given provision zone’s popula- tion were the same as that of the entire city. Comparing these two values (pa vs. pe) gives an insight into which age groups are underrepresented or overrepresented within the high, medium, or low zones regarding local green space availability, public green space accessibility, or overall UGS provision: pe = 0.01 pt ∙ r, where pe is the estimated population of a given age group within given overall UGS provision zone, local green space availabil- ity zone, or public green space accessibility zone (low, medium, or high); pt is the total population (sum of four age groups) with- in a given overall UGS provision zone, lo- cal green space availability zone, or public green space accessibility zone (low, medium, or high); and r is the proportion of a given age group within the total city population. We then calculated the percentage differ- ence (Eq. 2) between pa and pe by d = (pa – pe)/pa ∙ 100, where d is the percentage difference between the actual and expected number of residents within a given age group of the given overall UGS provision zone, local green space avail- ability zone, or public green space accessi- bility zone (low, medium, or high); pe is the estimated population of a given age group within the given overall UGS provision zone, local green space availability zone, or public green space accessibility zone (low, medium, or high); and pa is the actual population of a given age group within the given overall UGS provision zone, local green space avail- ability zone, or public green space accessibil- ity zone (low, medium, or high). Results Overall urban green space provision mapping On the basis of the local green space avail- ability map (Figure 5), we concluded that the overall state of local green space availability in Szeged is good. About 88 percent of the building plots have high local green space availability, while the average score is ~2.8. As expected, the largest concentration of building plots (classified as high, or scored as 3) are in the suburban areas, where de- tached houses are dominant. Building plots classified as medium (scored as 2) or low (scored as 1) are primarily in the city cen- tre, where the dense building coverage and high proportion of paved areas result in low NDVI values, and in housing areas where an otherwise acceptable amount of vegetation coverage has to satisfy the needs of a large residential population. Based on the results, the state of public green space accessibility appears to be lower throughout Szeged. Only ~16 percent of the building plots were identified as being part of the high public green space accessibility zones. In this case, the average score of the city was 1.6. Public green space accessibil- ity maps (Figures 6 and 7) indicate that the best public green space accessibility is within the more densely populated downtown of Szeged, where large public green spaces are more prominent, complementing the results of the local green space availability map. Low public green space accessibility building plots are primarily identified in the suburban areas, where public green spaces are less fre- quent compared to the city centre. Regarding the spatial characteristics of the overall UGS provision zones (Figure 8), it can be stated that in general, the overall UGS provision of Szeged is good. Approximately 38 percent of the building plots received high scores (5 or 6) as a result of the aggregation of their local green space availability and public green space accessibility scores. Additionally, another ~58 percent of the building plots were categorized as having medium overall (2) (1) 259Kolcsár, R.A. et al. Hungarian Geographical Bulletin 71 (2022) (3) 249–269. Fig. 5. Building-plot-scale local green space availability (square meters per capita), Szeged, Hungary. Fig. 6. Accessibility map, showing selected public green spaces with their designated size-based catchments in Szeged, Hungary. Kolcsár, R.A. et al. Hungarian Geographical Bulletin 71 (2022) (3) 249–269.260 Fig. 7. Building-plot-scale public green space accessibility, Szeged, Hungary. Fig. 8. Building-plot-scale overall urban green space provision map of Szeged, Hungary. 261Kolcsár, R.A. et al. Hungarian Geographical Bulletin 71 (2022) (3) 249–269. UGS provision (with a score of 4). The aver- age overall UGS provision score in Szeged was 4.5. Figure 8 also shows areas within the city that lack both local green space avail- ability and public green space accessibility. Building plots with a very low or low pub- lic green space accessibility score (Figure 9) need the most attention regarding green space development. Building plots in the very high overall UGS provision zones are generally found in the direct proximity of the public green spaces where part of the locally avail- able green space is the closest accessible pub- lic space itself. Population distribution assessment Results of the age-group analyses of differ- ent overall UGS provision zones (three-level classification) show that the vast majority of residents live in either medium or high over- all UGS provision zones. However, a notable number of people (~ 19,000) of the total pop- ulation (~144,000) live in areas categorized as low overall UGS provision zones (Figure 10). The population diagram (Figure 11) indi- cates that only a very small proportion of the population (~ 800 people) live in the green- deficit areas, for which the lowest possible overall UGS provision score (2) was applied. In contrast, the population of very high over- all UGS provision zones (score 6) is nomi- nally < 9,000. Both underrepresentation and overrepre- sentation are apparent when looking at the age groups (Figure 12). Compared to the age distribution of Szeged, the 0–18 age group is overrepresented in the areas with high lo- cal green space availability, implying that relatively speaking this age group has the best local green space availability in the city, Fig. 9. Building-plot-scale map of Szeged, Hungary, showing five urban green space provision zones. Kolcsár, R.A. et al. Hungarian Geographical Bulletin 71 (2022) (3) 249–269.262 while in an absolute sense, members of age group 41–60 are present within these zones in the largest number. In contrast, the 19–40 age group is highly underrepresented in the high local green space availability zones compared to the proportion of the other age groups. Similarly to the age group of 0–18, age groups of 41–60, and ≥ 61 are also over- represented in areas with high local green space availability. In many cases, we found that the popula- tion distribution from the public green space accessibility analyses show patterns opposite to the results of the local green space avail- ability mapping (Figure 13). The age group of 0–18 is strongly overrep- resented in areas of low public green space accessibility. In contrast, age groups of 19–40 and ≥ 61 are overrepresented in areas of high public green space accessibility. It is also noteworthy that more than a third of the to- tal population lives within low public green space accessibility zones. Overall UGS provi- sion analysis shows that while only a small proportion of the residents can be found in Fig. 10. Total population distribution of three overall urban green space provision categories for Szeged, Hungary. Fig. 12. Age-group distributions, Szeged, Hungary. A = Local green space availability zones; B = Percentage differences between actual and expected population. Fig. 11. Population distributions of five overall urban green space provision zones in Szeged, Hungary. Fig. 13. Age-group distributions, Szeged, Hungary. A = Public green space accessibility; B = Percentage differences between actual and expected population. 263Kolcsár, R.A. et al. Hungarian Geographical Bulletin 71 (2022) (3) 249–269. areas with low classification, their relative distribution within the zones favours the ≥ 61 age group (Figure 14). Within the low overall UGS provision zone, the 19–40 age group is the most overrepresented. In contrast, in the high overall UGS provision zone, every age group but the elderly (≥ 61) appears to be underrepresented compared to the actual age distribution of the entire population of Szeged (see Figure 14). Discussion As regards our findings, it can be stated, that based on our results the average level of over- all UGS provision in the city is slightly above medium. Our methodology identified areas within Szeged with a low score regarding both aspects (accessibility and availability) of UGS provision. Given that Kemperman, A.D.A.M. and Timmermans, H.J.P. (2007) argued that families with children are more likely to visit urban parks, the underrepre- sentation of age group 0–18 in public green space accessibility can be considered subop- timal. The elderly being favoured regarding UGS provision is not unique to Szeged, Wen, C. et al. (2020) drew similar conclusions in the case of Hannover, Germany. Good UGS provision is unquestionably beneficial for the elderly (especially for those ≥ 65), because UGSs provide them numerous physical and psychological benefits (Loukaitou-Sideris, A. et al. 2016), however, the UGS usage habits of this age group have their own peculiarities which need to be addressed in urban plan- ning. According to Kemperman, A.D.A.M. and Timmermans, H.J.P. (2007) for instance, UGS usage habits of seniors tend to be extreme, as they either visit urban parks or other public green spaces very frequently or almost never. Their willingness to visit UGSs is largely de- fined by the ecosystem services provided by these UGSs (e.g., where trees provide enough shading and sufficiently cool microclimate) as well as the infrastructure of both the UGSs and its neighbourhood (e.g., street furniture, well-maintained safe roads, etc.) (Kabisch, N. and Haase, D. 2014; Kántor, N. 2016; Loukaitou-Sideris, A. et al. 2016; Arnberger, A. et al. 2017; Artmann, M. et al. 2017; Wen, C. et al. 2020). Because public green space ac- cessibility favours seniors in Szeged, urban development should primarily focus on infra- structure development of these existing public green spaces as well as their neighbourhood to enhance the effective UGS usage by this age group (Levy-Storms, L. et al. 2018). The value of the overall UGS provision metric used in present the study could realis- tically be increased in two possible ways. The first way is to increase the vegetation cover- age (represented by NDVI values) within the 50-m proximity of the building plots where local green space availability is low. This can be done primarily via grassing or by bush or tree plantation, as well as by the creation of green roofs and green walls. The second possibility is to establish new public green spaces at the centre or otherwise in close proximity to building-plot clusters that have low UGS provision values. This method en- Fig. 14. Age-group distributions, Szeged, Hungary: (A) Overall urban green space provision zones; (B) Percentage difference between actual and expected population. Kolcsár, R.A. et al. Hungarian Geographical Bulletin 71 (2022) (3) 249–269.264 hances both public green space accessibility and local green space availability. This type of green space provision development might be most efficient in areas with large, but func- tionless green space patches, where urban parks or public gardens can be created via landscaping and expanding the area’s func- tions. Although the establishment of urban parks or other similar public green spaces is often limited by resources and available space, a growing body of literature demon- strates that pocket parks can be an effective solution to this problem, such as the work of Kerishnan, P.B. and Maruthaveeran, S. (2021), Naghibi, M. et al. (2021) or Rosso, F. et al. (2022). Our methodology also enables identify- ing age groups that are underrepresented in high UGS provision areas (e.g., age group of 0–18 years in the case of Szeged) to prioritize UGS development in low UGS provision ar- eas where the concentration of the age group is higher. Conclusions In this study, we calculated the building- plot-scale UGS provision for Szeged, Hun- gary with a methodology incorporating two of the prior-established UGS provision levels: availability and accessibility (Biernacka, M. and Kronenberg, J. 2019). In Szeged, local green space availability is higher in the pe- ripheral, single-family housing areas, while public green space accessibility is higher in the densely populated city centre. These two metrics complement each other, provid- ing good overall UGS provision conditions throughout the residential areas of the entire city. The majority of the residents live in ar- eas with medium or high UGS provision. All the age groups except the elderly (≥ 61) are underrepresented in areas with high overall UGS provision. This is especially true of the 0–18 age group. Our methodology proved to be an ade- quate tool to delineate zones within Szeged with insufficient UGS provision, as well as identify disadvantageous age groups within these areas. We hope that the present meth- odology and our results will be applicable in other cities for high-resolution preliminary UGS provision assessments as well. There are, however, certain limitations that need to be addressed in the future. Our methodology used numerous parameters based on (or derived from) internationally accepted thresholds (e.g., category-limit val- ues and applied buffer-zone radius for local green space availability, catchment sizes for public green space accessibility, etc.). In the future, more robust inferences could be made of the UGS provision by sensitivity analyses to different parameters. Future studies should also aim to inves- tigate the UGS provision of different age groups separately, applying parameters that better reflect the examined focus group. For example, the public green space accessibility assessment of the elderly, who are generally less mobile than the younger age groups, should be carried out with a lower aver- age walking velocity (compared to the 5 km per hour walking speed used in the present study). In future studies, vehicle-based acces- sibility mapping besides the walking-based method should also be considered. A clear limitation of the broad applicabil- ity of our framework is that data at a similar resolution might not be available for other study areas. Knowing such limitations, cer- tain input data can be replaced by more broadly available but less-detailed data, such as population attributes from the 2012 and 2018 European Environmental Agencies Urban Atlas land use and land cover data- bases (Copernicus 2018; Kolcsár, R.A. et al. 2021). A different, raster-based method for the identification of public green spaces, e.g., methodologies proposed by Huang, B. et al. (2018), and Bui, D.H. and Mucsi, L. (2021), should also be tested in the future. Acknowledgements: Project no. TKP2021-NVA-09 has been implemented with the support provided by the Ministry of Innovation and Technology of Hungary from the National Research, Development and 265Kolcsár, R.A. et al. Hungarian Geographical Bulletin 71 (2022) (3) 249–269. Innovation Fund, financed under the TKP2021-NVA funding scheme. The study was also supported by NTP-NFTÖ-20 of the Ministry of Human Capacities, Hungary and the Hungarian Ministry of Interior. We wish to show our appreciation for the Hungarian Ministry of Interior for providing us with the detailed population data of Szeged, without which the present study would not have been possible; for Enago for the grammatical and stylistic improvements on our manuscript; and for the Department of Climatology and Landscape Ecology for their contribution to our work via Szeged’s orthophoto footage. We are also thankful for the kind work of the anonym reviewer, who supplied us with useful comments on an earlier version of this paper; as well as our colleague, who took her time to do the proofreading on the revised version of our manuscript. Finally, wish to extend our special thanks to Máté Tóth, for his kind professional advices regarding some of the spatial data that were used in the research. REFERENCES Aquino, D. do N., Neto, O.C. da R., Moreira, M.A., Teixeira, A. dos S. and de Andrade, E.M. 2018. Use of remote sensing to identify areas at risk of degradation in the semi-arid region. Revista Ciencia Agronomica 49. (3): 420–429. Available at https://doi. org/10.5935/1806-6690.20180047 Arnberger, A., Allex, B., Eder, R., Ebenberger, M., Wanka, A., Kolland, F., Wallner, P. and Hutter, H.P. 2017. Elderly resident’s uses of and preferences for urban green spaces during heat periods. Urban Forestry and Urban Greening 21. 102–115. Available at https://doi.org/10.1016/j.ufug.2016.11.012 Artmann, M., Chen, X., Iojă, C., Hof, A., Onose, D., Poniży, L., Lamovšek, A.Z. and Breuste, J. 2017. The role of urban green spaces in care facilities for elderly people across European cities. Urban Forestry and Urban Greening 27. 203–213. Available at https://doi.org/10.1016/j.ufug.2017.08.007 Ayala-Azcárraga, C., Diaz, D. and Zambrano, L. 2019. Characteristics of urban parks and their relation to user well-being. Landscape and Urban Planning 189. 27–35. Available at https://doi. org/10.1016/j.landurbplan.2019.04.005 Bahrini, F., Bell, S. and Mokhtarzadeh, S. 2017. The relationship between the distribution and use patterns of parks and their spatial accessibility at the city level: A case study from Tehran, Iran. Urban Forestry and Urban Greening 27. 332–342. Available at https://doi.org/10.1016/j.ufug.2017.05.018 Barbosa, O., Tratalos, J.A., Armsworth, P.R., Davies, R.G., Fuller, R.A., Johnson, P. and Gaston, K.J. 2007. Who benefits from access to green space? A case study from Sheffield, UK. Landscape and Urban Planning 83. 187–195. Available at https://doi. org/10.1016/j.landurbplan.2007.04.004 Belügyminisztérium 2019. Hungarian Ministry of the Interior. Database of the Hungarian Ministry of the Interior 2018. Budapest, Hungarian Ministry of Interior. Biernacka, M. and Kronenberg, J. 2019. Urban green space availability, accessibility and attractiveness, and the delivery of ecosystem services. Cities and the Environment 12. (1):5. Available at https://digi- talcommons.lmu.edu/cate/vol12/iss1/5 Biernacka, M., Kronenberg, J. and Łaszkiewicz, E. 2020. An integrated system of monitoring the availability, accessibility and attractiveness of urban parks and green squares. Applied Geography 116. 102152. Available at https://doi.org/10.1016/j. apgeog.2020.102152 Blaschke, T. and Kovács-Győri, A. 2020. Earth observation to substantiate the sustainable develop- ment goal 11: Practical considerations and experi- ences from Austria. The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, Vol. XLIII-B4-2020, XXIV ISPRS Congress. Available at https://doi. org/10.5194/isprs-archives-XLIII-B4-2020-769-2020 Bok, J. and Kwon, Y. 2016. Comparable measures of accessibility to public transport using the general transit feed specification. Sustainability (Switzerland) 8. (3): 1–224. Available at https://doi.org/10.3390/ su8030224 Bozkurt, M. 2021. Metropolitan children’s physical fitness: The relationship between overweight and obesity prevalence, socioeconomic status, urban green space access, and physical activity. Urban Forestry and Urban Greening 64. 127272. Available at https://doi.org/10.1016/j.ufug.2021.127272 Braquinho, C., Cvejić, R., Eler, K., Gonzales, P., Haase, D., Hansen, R., Kabisch, N., Lorance Rall, E., Niemela, J., Pauleit, S., Pintar, M., Lafortezza, R., Santos, A., Strohbach, M., Vierikko, K. and Železnikar, Š. 2015. A Typology of Urban Green Spaces, Ecosystem Provisioning Services and Demands. Green Surge, Report D3.1. EU FP7. Available at https://assets.centralparknyc. org/pdfs/institute/p2p-upelp/1.004_Greensurge_ A+Typology+of+Urban+Green+Spaces.pdf Bui, D.H. and Mucsi, L. 2021. From land cover map to land use map: A combined pixel-based and object- based approach using multi-temporal Landsat data, a random forest classifier, and decision rules. Remote Sensing 13. (9): 1700. Available at https://doi. org/10.3390/rs13091700 Chênes, C., Giuliani, G. and Ray, N. 2021. Modelling physical accessibility to public green spaces in Switzerland to support the SDG11. Geomatics 1. (4): 383–398. Available at https://doi.org/10.3390/ geomatics1040022 Cheng, Y., Zhang, J., Wei, W. and Zhao, B. 2021. Effects of urban parks on residents’ expressed https://doi.org/10.5935/1806-6690.20180047 https://doi.org/10.5935/1806-6690.20180047 https://doi.org/10.1016/j.ufug.2016.11.012 https://doi.org/10.1016/j.landurbplan.2019.04.005 https://doi.org/10.1016/j.landurbplan.2019.04.005 https://doi.org/10.1016/j.landurbplan.2007.04.004 https://doi.org/10.1016/j.landurbplan.2007.04.004 https://digitalcommons.lmu.edu/cate/vol12/iss1/5 https://digitalcommons.lmu.edu/cate/vol12/iss1/5 https://doi.org/10.1016/j.apgeog.2020.102152 https://doi.org/10.1016/j.apgeog.2020.102152 https://doi.org/10.5194/isprs-archives-XLIII-B4-2020-769-2020 https://doi.org/10.5194/isprs-archives-XLIII-B4-2020-769-2020 https://doi.org/10.3390/su8030224 https://doi.org/10.3390/su8030224 https://assets.centralparknyc.org/pdfs/institute/p2p-upelp/1.004_Greensurge_A+Typology+of+Urban+Green+Spaces.pdf https://assets.centralparknyc.org/pdfs/institute/p2p-upelp/1.004_Greensurge_A+Typology+of+Urban+Green+Spaces.pdf https://assets.centralparknyc.org/pdfs/institute/p2p-upelp/1.004_Greensurge_A+Typology+of+Urban+Green+Spaces.pdf https://doi.org/10.3390/rs13091700 https://doi.org/10.3390/rs13091700 https://doi.org/10.3390/geomatics1040022 https://doi.org/10.3390/geomatics1040022 Kolcsár, R.A. et al. Hungarian Geographical Bulletin 71 (2022) (3) 249–269.266 happiness before and during the COVID-19 pan- demic. Landscape and Urban Planning 212. 104118. Available at https://doi.org/10.1016/j.landurb- plan.2021.104118 CICES 2022. Common International Classification of Ecosystem Services. CICES Version 5.1. Copenhagen, European Environment Agency. Copernicus 2018. Urban Atlas 2018. Copernicus Land Monitoring Service, EU. Csete, Á.K. and Gulyás, Á. 2021. Green infrastruc- ture-based hydrological modelling, a comparison between different urban districts, through the case of Szeged, Hungary. Hungarian Geographical Bulletin 70. (4): 353–368. Available at https://doi. org/10.15201/hungeobull.70.4.5 Csete, Á.K., Kolcsár, R.A. and Gulyás, Á. 2021. Rainwater harvesting potential and vegetation irrigation assessment derived from building data- based hydrological modelling through the case study of Szeged. Carpathian Journal of Earth and Environmental Sciences 16. (2): 469–482. Available at https://doi.org/10.26471/cjees/2021/016/192 Csomós, G., Farkas, J.Z. and Kovács, Z. 2020. Access to urban green spaces and environmental inequal- ity in post-socialist cities. Hungarian Geographical Bulletin 69. (2): 191–207. Available at https://doi. org/10.15201/hungeobull.69.2.7 Csomós, G., Farkas, Z.J., Kolcsár, R.A., Szilassi, P. and Kovács, Z. 2021. Measuring socio-economic disparities in green space availability in post-social- ist cities. Habitat International 117. 102434. Available at https://doi.org/10.1016/j.habitatint.2021.102434 Edwards, D., Elliott, A., Hislop, M., Martin, S., Morris, J., O’Brien, L., Peace, A., Sarajevs, V., Serrand, M. and Valatin, G. 2009. A Valuation of the Economic and Social Contribution of Forestry for People in Scotland. Edinburgh, Forestry Commission Scotland. Available at https://cdn.forestresearch. gov.uk/2009/01/fcrp101.pdf Ekkel, E.D. and de Vries, S. 2017. Nearby green space and human health: Evaluating accessibility metrics. Landscape and Urban Planning 157. 214–220. Available at https://doi.org/10.1016/j.landurb- plan.2016.06.008 E S R I 2018. Environmental Systems Research Institute. ArcGIS Network Analyst. Redlands, California, ESRI. Fehér, É. 2018. Szeged megyei jogú város építési szabá- lyzatának módosítása a lakás rendeltetési egységekhez tartozó telekhányadok meghatározására vonatkozóan (Modification of regulation of constructions for county seat Szeged retating to building plot rate of flat units). Szeged, Szeged Megyei Jogú Város Önkormányzata. Available at https://docplayer. hu/105688815-Szeged-megyei-jogu-varos-epitesi- szabalyzatanak.html Fuller, R.A. and Gaston, K.J. 2009. The scaling of green space coverage in European cities. Biology Letters 5. (3): 352–355. Available at https://doi. org/10.1098/rsbl.2009.0010 Gál, T., Skarbit, N. and Unger, J. 2016. Urban heat island patterns and their dynamics based on an urban climate measurement network. Hungarian Geographical Bulletin 65. (2): 105–116. Available at https://doi.org/10.15201/hungeobull.65.2.2 Gehl, J. 2010. Cities for People. Washington, D. C., Island Press. Geofabrik 2018. Geofabrik OpenStreetMap Data Extracts. Access date: 2018.12.05. Geofabrik 2021. Geofabrik OpenStreetMap Data Extracts Access date: 2021.02.05. Hare, T.S. and Barcus, H.R. 2007. Geographical acces- sibility and Kentucky’s heart-related hospital ser- vices. Applied Geography 27. (3–4): 181–205. Available at https://doi.org/10.1016/j.apgeog.2007.07.004 Henits, L., Jürgens, C. and Mucsi, L. 2016. Seasonal multitemporal land-cover classification and change detection analysis of Bochum, Germany, using mul- titemporal Landsat TM data. International Journal of Remote Sensing 37. (15): 3439–3454. Available at https://doi.org/10.1080/01431161.2015.1125558 Herbel, I., Croitoru, A.E., Rus, I., Harpa, G.V. and Ciupertea, A.F. 2016. Detection of atmospheric urban heat island through direct measurements in Cluj-Napoca city, Romania. Hungarian Geographical Bulletin 65. (2): 117–128. Available at https://doi. org/10.15201/hungeobull.65.2.3 Huang, B., Zhao, B. and Song, Y. 2018. Urban land- use mapping using a deep convolutional neural network with high spatial resolution multispec- tral remote sensing imagery. Remote Sensing of Environment 214. 73–86. Available at https://doi. org/10.1016/J.RSE.2018.04.050 James, P., Tzoulas, K., Adams, M.D., Barber, A., Box, J., Breuste, J., Elmqvist, T., Frith, M., Gordon, C., Greening, K.L., Handley, J., Haworth, S., Kazmierczak, A. E., Johnston, M., Korpela, K., Moretti, M., Niemelä, J., Pauleit, S., Roe, M.H., Sadler, J.P. and Ward Thompson, C. 2009. Towards an integrated understanding of green space in the European built environment. Urban Forestry and Urban Greening 8. (2): 65–75. Available at https:// doi.org/10.1016/j.ufug.2009.02.001 Jiang, S., Alves, A., Rodrigues, F., Ferreira, J. and Pereira, F.C. 2015. Mining point-of-interest data from social networks for urban land use classifica- tion and disaggregation. Computers, Environment and Urban Systems 53. 36–46. Available at https://doi. org/10.1016/j.compenvurbsys.2014.12.001 Kabisch, N. and Haase, D. 2014. Green justice or just green? Provision of urban green spaces in Berlin, Germany. Landscape and Urban Planning 122. 129–139. Available at https://doi.org/10.1016/j. landurbanplan.2013.11.016 Kabisch , N., Strohbach , M., Haase , D. and Kronenberg, J. 2016. Urban green space avail- https://doi.org/10.1016/j.landurbplan.2021.104118 https://doi.org/10.1016/j.landurbplan.2021.104118 https://doi.org/10.15201/hungeobull.70.4.5 https://doi.org/10.15201/hungeobull.70.4.5 https://doi.org/10.26471/cjees/2021/016/192 https://doi.org/10.15201/hungeobull.69.2.7 https://doi.org/10.15201/hungeobull.69.2.7 https://doi.org/10.1016/j.habitatint.2021.102434 https://cdn.forestresearch.gov.uk/2009/01/fcrp101.pdf https://cdn.forestresearch.gov.uk/2009/01/fcrp101.pdf https://doi.org/10.1016/j.landurbplan.2016.06.008 https://doi.org/10.1016/j.landurbplan.2016.06.008 https://docplayer.hu/105688815-Szeged-megyei-jogu-varos-epitesi-szabalyzatanak.html https://docplayer.hu/105688815-Szeged-megyei-jogu-varos-epitesi-szabalyzatanak.html https://docplayer.hu/105688815-Szeged-megyei-jogu-varos-epitesi-szabalyzatanak.html https://doi.org/10.1098/rsbl.2009.0010 https://doi.org/10.1098/rsbl.2009.0010 https://doi.org/10.15201/hungeobull.65.2.2 https://doi.org/10.1016/j.apgeog.2007.07.004 https://doi.org/10.1080/01431161.2015.1125558 https://doi.org/10.15201/hungeobull.65.2.3 https://doi.org/10.15201/hungeobull.65.2.3 https://doi.org/10.1016/J.RSE.2018.04.050 https://doi.org/10.1016/J.RSE.2018.04.050 https://doi.org/10.1016/j.ufug.2009.02.001 https://doi.org/10.1016/j.ufug.2009.02.001 https://doi.org/10.1016/j.compenvurbsys.2014.12.001 https://doi.org/10.1016/j.compenvurbsys.2014.12.001 https://doi.org/10.1016/j.landurbplan.2013.11.016 https://doi.org/10.1016/j.landurbplan.2013.11.016 267Kolcsár, R.A. et al. Hungarian Geographical Bulletin 71 (2022) (3) 249–269. ability in European cities. Ecological Indicators 70. 586–596. Available at https://doi.org/10.1016/j. ecolind.2016.02.029 Kántor, N. 2016. Differences between the evaluation of thermal environment in shaded and sunny position. Hungarian Geographical Bulletin 65. (2): 139–153. Available at https://doi.org/10.15201/ hungeobull.65.2.5 Kemperman, A.D.A.M. and Timmermans, H.J.P. 2007. Heterogeneity in urban park use of ag- ing visitors: A latent class analysis. Leisure Sciences 28. (1): 57–71. Available at https://doi. org/10.1080/01490400500332710 Kerishnan, P.B. and Maruthaveeran, S. 2021. Factors contributing to the usage of pocket parks – A review of the evidence. Urban Forestry and Urban Greening 58. 126985. Available at https://doi.org/10.1016/j. ufug.2021.126985 Kolcsár, R.A., Csikós, N. and Szilassi, P. 2021. Testing the limitations of buffer zones and Urban atlas pop- ulation data in urban green space provision analyses through the case study of Szeged, Hungary. Urban Forestry and Urban Greening 57. 126942. Available at https://doi.org/10.1016/j.ufug.2020.126942 Koprowska, K., Łaszkiewicz, E., Kronenberg, J. and Marcińczak, S. 2018. Subjective perception of noise exposure in relation to urban green space availability. Urban Forestry and Urban Greening 31. 93–102. Available at https://doi.org/10.1016/j. ufug.2018.01.018 Kothencz, G., Kolcsár, R., Cabrera-Barona, P. and Szilassi, P. 2017. Urban green space perception and its contribution to well-being. International Journal of Environmental Research and Public Health 14. (7): 766. Available at https://doi.org/10.3390/ijerph14070766 Kovács-Győri, A., Ristea, A., Kolcsar, R., Resch, B., Crivellari, A. and Blaschke, T. 2018. Beyond spa- tial proximity-classifying parks and their visitors in london based on spatiotemporal and sentiment analysis of twitter data. ISPRS International Journal of Geo-Information 7. (9): 378. Available at https://doi. org/10.3390/ijgi7090378 Kraemer, R. and Kabisch, N. 2021. Parks in context: advancing citywide spatial quality assessments of urban green spaces using fine-scaled indicators. Ecology and Society 26. (2): 45. Available at https:// doi.org/10.5751/es-12485-260245 Kraft, S. 2016. Measuring and modelling the spa- tial accessibility of public transport stops in GIS. Hungarian Geographical Bulletin 65. (1): 57–69. Available at https://doi.org/10.15201/hungeob- ull.65.1.5 Kronenberg, J. 2015. Why not to green a city? Institutional barriers to preserving urban ecosystem services. Ecosystem Services 12. 218–227. Available at https://doi.org/10.1016/j.ecoser.2014.07.002 KSH 2021. Hungarian Statistical Office. Demography, 2020. Budapest, Hungarian Central Statistical Office. Kwan, M.P. and Weber, J. 2008. Scale and accessibility: Implications for the analysis of land use-travel inter- action. Applied Geography 28. (2): 110–123. Available at https://doi.org/10.1016/j.apgeog.2007.07.002 Le Texier, M., Schiel, K. and Caruso, G. 2018. The provision of urban green space and its accessibility: Spatial data effects in Brussels. PLoS ONE 13. (10): e0204684. Available at https://doi.org/10.1371/jour- nal.pone.0204684 Lechner Tudásközpont 2015. Data from the Lechner Knowledge Center (former Institute of Geodesy, Cartography and Remote Sensing), Budapest. Levy-Storms, L., Chen, L. and Loukaitou-Sideris, A. 2018. Older adults’ needs and preferences for open space and physical activity in and near parks: A sys- tematic review. Journal of Aging and Physical Activity 26. (4): 682–696. Available at https://doi.org/10.1123/ japa.2016-0354 Loukaitou-Sideris, A., Levy-Storms, L., Chen, L. and Brozen, M. 2016. Parks for an aging population: Needs and preferences of low-income seniors in Los Angeles. Journal of the American Planning Association 82. (3): 236–251. Available at https://doi.org/10.108 0/01944363.2016.1163238 Makovníková, J., Pálka, B., Širáň, M., Kizeková, M. and Kanianska, R. 2019. The potential of regulating ecosystem service – filtering potential for inorganic pollutants – supplied by soils of Slovakia. Hungarian Geographical Bulletin 68. (2): 177–185. Available at https://doi.org/10.15201/hungeobull.68.2.5 Maryanti, M.R., Khadijah, H., Uzair, A.M., Megat, M.A.R. and Ghazali, M. 2016. The urban green space provision using the standards approach: issues and challenges of its implementation in Malaysia. Sustainable Development and Planning 8. 369–379. Available at https://doi.org/10.2495/sdp160311 McGrail, M.R. and Humphreys, J.S. 2009. Measuring spatial accessibility to primary care in rural areas: Improving the effectiveness of the two-step float- ing catchment area method. Applied Geography 29. (4): 533–541. Available at https://doi.org/10.1016/j. apgeog.2008.12.003 MEA 2005. Millennium Ecosystem Assessment. World Resources Institute, UNEP. Available at https:// www.millenniumassessment.org/en/Index-2.html Mora-Garcia, R.T., Marti-Ciriquian, P., Perez- Sanchez, R. and Cespedes-Lopez, M.F. 2018. A com- parative analysis of manhattan, euclidean and network distances. Why are network distances more useful to urban professionals? International Multidisciplinary Scientific GeoConference Surveying Geology and Mining Ecology Management. SGEM Scientific eLibrary. Available at https://www.sgem.org/index. php/elibrary?view=publication&task=show&id=602 Morar, T., Radoslav, R., Spiridon, L. and Păcurar, L. 2014. Assessing pedestrian accessibility to green space using GIS. Transylvanian Review of Administrative Sciences 10. (42): 116–139. https://doi.org/10.1016/j.ecolind.2016.02.029 https://doi.org/10.1016/j.ecolind.2016.02.029 https://doi.org/10.15201/hungeobull.65.2.5 https://doi.org/10.15201/hungeobull.65.2.5 https://doi.org/10.1080/01490400500332710 https://doi.org/10.1080/01490400500332710 https://doi.org/10.1016/j.ufug.2021.126985 https://doi.org/10.1016/j.ufug.2021.126985 https://doi.org/10.1016/j.ufug.2020.126942 https://doi.org/10.1016/j.ufug.2018.01.018 https://doi.org/10.1016/j.ufug.2018.01.018 https://doi.org/10.3390/ijerph14070766 https://doi.org/10.3390/ijgi7090378 https://doi.org/10.3390/ijgi7090378 https://doi.org/10.5751/es-12485-260245 https://doi.org/10.5751/es-12485-260245 https://doi.org/10.15201/hungeobull.65.1.5 https://doi.org/10.15201/hungeobull.65.1.5 https://doi.org/10.1016/j.ecoser.2014.07.002 https://doi.org/10.1016/j.apgeog.2007.07.002 https://doi.org/10.1371/journal.pone.0204684 https://doi.org/10.1371/journal.pone.0204684 https://doi.org/10.1123/japa.2016-0354 https://doi.org/10.1123/japa.2016-0354 https://doi.org/10.1080/01944363.2016.1163238 https://doi.org/10.1080/01944363.2016.1163238 https://doi.org/10.15201/hungeobull.68.2.5 https://doi.org/10.2495/sdp160311 https://doi.org/10.1016/j.apgeog.2008.12.003 https://doi.org/10.1016/j.apgeog.2008.12.003 https://www.millenniumassessment.org/en/Index-2.html https://www.millenniumassessment.org/en/Index-2.html https://www.sgem.org/index.php/elibrary?view=publication&task=show&id=602 https://www.sgem.org/index.php/elibrary?view=publication&task=show&id=602 Kolcsár, R.A. et al. Hungarian Geographical Bulletin 71 (2022) (3) 249–269.268 Mucsi, L., Liska, C.M., Henits, L., Tobak, Z., Csendes, B. and Nagy, L. 2017. The evaluation and applica- tion of an urban land cover map with image data fusion and laboratory measurements. Hungarian Geographical Bulletin 66. (2): 145–156. Available at https://doi.org/10.15201/hungeobull.66.2.4 Naghibi, M., Faizi, M. and Ekhlassi, A. 2021. Design possibilities of leftover spaces as a pocket park in relation to planting enclosure. Urban Forestry and Urban Greening 64. 127273. Available at https://doi. org/10.1016/j.ufug.2021.127273 Neuvonen, M., Sievänen, T., Tönnes, S. and Koskela, T. 2007. Access to green areas and the frequency of visits – A case study in Helsinki. Urban Forestry and Urban Greening 6. (4): 235–247. Available at https:// doi.org/10.1016/j.ufug.2007.05.003 Oh, K. and Jeong, S. 2007. Assessing the spatial dis- tribution of urban parks using GIS. Landscape and Urban Planning 82. (1–2): 25–32. Available at https:// doi.org/10.1016/j.landurbplan.2007.01.014 Pafi, M., Siragusa, A., Ferri, S. and Halkia, S. 2016. Measuring the accessibility of urban green areas: A comparison of the Green ESM with other datasets in four European cities. JRC Publications repository. Luxembourg, Publications Office of the European Union. Available at https://publications.jrc.ec.euro- pa.eu/repository/handle/JRC102525 Park, K. 2017. Psychological park accessibility: a systematic literature review of perceptual compo- nents affecting park use. Landscape Research 42. (5): 508–520. Available at https://doi.org/10.1080/0142 6397.2016.1267127 Poleman, H. 2018. A walk to the park? Assessing access to green areas in Europe’s cities. Update using completed Copernicus urban atlas data. WP 01/2018. European Commission. Available at https://ec.europa.eu/ regional_policy/sources/docgener/work/2018_01_ green_urban_area.pdf Pongrácz, R., Bartholy, J., Dezső, Z. and Dian, C. 2016. Analysis of the air temperature and relative humidity measurements in Budapest-Ferencváros. Hungarian Geographical Bulletin 65. (2): 93–103. Available at https://doi.org/10.15201/hungeob- ull.65.2.1 Razak, M.A.W.A., Othman, N. and Nazir, N.N.M. 2016. Connecting people with nature: Urban park and human well-being. Procedia – Social and Behavioral Sciences 222. 476–484. Available at https:// doi.org/10.1016/J.SBSPRO.2016.05.138 Rossi, S.D., Byrne, J.A. and Pickering, C.M. 2015. The role of distance in peri-urban national park use: Who visits them and how far do they travel? Applied Geography 63. 77–88. Available at https://doi. org/10.1016/J.APGEOG.2015.06.008 Rosso, F., Cappa, F., Spitzmiller, R. and Ferrero, M. 2022. Pocket parks towards more sustainable cities. Architectural, environmental, managerial and legal considerations towards an integrated framework: A case study in the Mediterranean re- gion. Environmental Challenges 7. 100402. Available at https://doi.org/10.1016/j.ufug.2021.126985 Rupprecht, D.C., Byrne, J.A., Ueda, H., and Lo, A.Y. 2015. ‘It’s real, not fake like a park’: Residents’ per- ception and use of informal urban green-space in Brisbane, Australia and Sapporo, Japan. Landscape and Urban Planning 143. 205–218. Available at https://doi.org/10.1016/j.landurbplan.2015.07.003 Schipperijn, J., Stigsdotter, U.K., Randrup, T.B. and Troelsen, J. 2010. Influences on the use of urban green space – A case study in Odense, Denmark. Urban Forestry and Urban Greening 9. (1): 25–32. Available at https://doi.org/10.1016/j. ufug.2009.09.002 Shahid, R., Bertazzon, S., Knudtson, M.L. and Ghali, W.A. 2009. Comparison of distance measures in spatial analytical modeling for health service plan- ning. BMC Health Services Research 9. 200. Available at https://doi.org/10.1186/1472-6963-9-200 Stanners, D. and Bourdeau, P. (eds.) 1995. Europe’s Environment: the Dobris Assessment. Specially prepared for the presentation of the European Environment Agency . London, Earthscan Publications. Available at https://www.eea.europa.eu/publica- tions/92-827-5122-8/page001.html Stessens, P., Khan, A.Z., Huysmans, M. and Canters, F. 2017. Analysing urban green space accessibility and quality: A GIS-based model as spatial decision support for urban ecosystem services in Brussels. Ecosystem Services 28. 328–340. Szilassi, P., Breuste, J., Kolcsár, A.R. and Aigner, G. 2020. Mobile application-based field survey as possible tool for investigating visitors’ perception and preferences of the vegetation. In Making Green Cities. Eds.: Breuste, J., Artmann, M., Ioja, C. and Quareshi, S., Cham, Springer, 459–473. Takács, Á., Kovács, A., Kiss, M., Gulyás, Á. and Kántor, N. 2016. Study on the transmissivity characteristics of urban trees in Szeged, Hungary. Hungarian Geographical Bulletin 65. (2): 155–167. Available at https://doi.org/10.15201/hungeob- ull.65.2.6 TEEB 2022. The Economics of Ecosystems and Biodiversity Project. United Nations Environment Programme. Geneva, Switzerland, UNEP FI. USGS 2018. United States Geological Survey. NDVI, the Foundation for Remote Sensing Phenology U.S. Geological Survey. Uzzoli, A., Egri, Z., Szilágyi, D. and Pál, V. 2020. Does better availability mean better accessibility? Spatial inequalities in the care of acute myocar- dial infarction in Hungary. Hungarian Geographical Bulletin 69. (4): 401–418. Available at https://doi. org/10.15201/hungeobull.69.4.5 Van den Bosch, C.K. 2021. Promoting health and wellbe- ing through urban forests – Introducing the 3-30-300 rule. IUCN Urban Alliance publication. Available https://doi.org/10.15201/hungeobull.66.2.4 https://doi.org/10.1016/j.ufug.2021.127273 https://doi.org/10.1016/j.ufug.2021.127273 https://doi.org/10.1016/j.ufug.2007.05.003 https://doi.org/10.1016/j.ufug.2007.05.003 https://doi.org/10.1016/j.landurbplan.2007.01.014 https://doi.org/10.1016/j.landurbplan.2007.01.014 https://publications.jrc.ec.europa.eu/repository/handle/JRC102525 https://publications.jrc.ec.europa.eu/repository/handle/JRC102525 https://doi.org/10.1080/01426397.2016.1267127 https://doi.org/10.1080/01426397.2016.1267127 https://ec.europa.eu/regional_policy/sources/docgener/work/2018_01_green_urban_area.pdf https://ec.europa.eu/regional_policy/sources/docgener/work/2018_01_green_urban_area.pdf https://ec.europa.eu/regional_policy/sources/docgener/work/2018_01_green_urban_area.pdf https://doi.org/10.15201/hungeobull.65.2.1 https://doi.org/10.15201/hungeobull.65.2.1 https://doi.org/10.1016/J.SBSPRO.2016.05.138 https://doi.org/10.1016/J.SBSPRO.2016.05.138 https://doi.org/10.1016/J.APGEOG.2015.06.008 https://doi.org/10.1016/J.APGEOG.2015.06.008 https://doi.org/10.1016/j.ufug.2021.126985 https://doi.org/10.1016/j.landurbplan.2015.07.003 https://doi.org/10.1016/j.ufug.2009.09.002 https://doi.org/10.1016/j.ufug.2009.09.002 https://doi.org/10.1186/1472-6963-9-200 https://www.eea.europa.eu/publications/92-827-5122-8/page001.html https://www.eea.europa.eu/publications/92-827-5122-8/page001.html https://doi.org/10.15201/hungeobull.65.2.6 https://doi.org/10.15201/hungeobull.65.2.6 https://doi.org/10.15201/hungeobull.69.4.5 https://doi.org/10.15201/hungeobull.69.4.5 269Kolcsár, R.A. et al. Hungarian Geographical Bulletin 71 (2022) (3) 249–269. at https://iucnurbanalliance.org/promoting-health- and-wellbeing-through-urban-forests-introducing- the-3-30-300-rule/ Weldon, S., Bailey, C. and O’Brien, L. 2007. New pathways for health and well-being in Scotland: Research to understand and overcome barriers to ac- cessing woodlands. Edinburgh, Forestry Commission Scotland. Available at https://cdn.forestresearch. gov.uk/2007/11/new_pathways_to_health_sum- mary.pdf Wen, C., Albert, C. and von Haaren, C. 2020. Equality in access to urban green spaces: A case study in Hannover, Germany, with a focus on the elderly population. Urban Forestry and Urban Greening 55. 126820. Available at https://doi. org/10.1016/j.ufug.2020.126820 Wolch, J.R., Byrne, J. and Newell, J.P. 2014. Urban green space, public health, and environmental justice: The challenge of making cities ‘just green enough’. Landscape and Urban Planning 125. 234–244. Available at https://doi.org/10.1016/j.landurb- plan.2014.01.017 WHO 2017. Urban planning, environment and health: from evidence to policy action. WHO report. New York, UN World Health Organization. WHO Regional Office for Europe 2017. Urban green spaces: a brief for action. WHO report. New York, UN World Health Organization.. Wright, W.H.E., Zarger, R.K. and Mihelcic, J.R. 2012. Accessibility and usability: Green space prefer- ences, perceptions, and barriers in a rapidly urban- izing city in Latin America. Landscape and Urban Planning 107. (3): 272–282. Available at https://doi. org/10.1016/j.landurbplan.2012.06.003 Wu, L. and Kim, S.K. 2021. Exploring the equality of accessing urban green spaces: A comparative study of 341 Chinese cities. Ecological Indicators 121. 107080. Available at https://doi.org/10.1016/j. ecolind.2020.107080 Wüstemann, H., Kalisch, D. and Kolbe, J. 2017. Access to urban green space and environmental inequalities in Germany. Landscape and Urban Planning 164. 124–131. Available at https://doi. org/10.1016/j.landurbplan.2017.04.002 Yang, L., Liu, J., Liang, Y., Lu, Y. and Yang, H. 2021. Spatially varying effects of street greenery on walk- ing time of older adults. ISPRS International Journal of Geo-Information 10. (9): 596. Available at https:// doi.org/10.3390/ijgi10090596 You, H. 2016. Characterizing the inequalities in urban public green space provision in Shenzhen, China. Habitat International 56. 176–180. Available at https:// doi.org/10.1016/j.habitatint.2016.05.006 Zepp, H., Gross, L. and Inostroza, L. 2020. And the winner is? Comparing urban green space provision and accessibility in eight European metropolitan areas using a spatially explicit approach. Urban Forestry and Urban Greening 49. 126603. Available at https://doi.org/10.1016/j.ufug.2020.126603 Zhang, X., Du, S. and Wang, Q. 2017. Hierarchical semantic cognition for urban functional zones with VHR satellite images and POI data. ISPRS Journal of Photogrammetry and Remote Sensing 132. 170–184. Available at https://doi.org/10.1016/j. isprsjprs.2017.09.007 https://iucnurbanalliance.org/promoting-health-and-wellbeing-through-urban-forests-introducing-the-3-30-300-rule/ https://iucnurbanalliance.org/promoting-health-and-wellbeing-through-urban-forests-introducing-the-3-30-300-rule/ https://iucnurbanalliance.org/promoting-health-and-wellbeing-through-urban-forests-introducing-the-3-30-300-rule/ https://cdn.forestresearch.gov.uk/2007/11/new_pathways_to_health_summary.pdf https://cdn.forestresearch.gov.uk/2007/11/new_pathways_to_health_summary.pdf https://cdn.forestresearch.gov.uk/2007/11/new_pathways_to_health_summary.pdf https://doi.org/10.1016/j.ufug.2020.126820 https://doi.org/10.1016/j.ufug.2020.126820 https://doi.org/10.1016/j.landurbplan.2014.01.017 https://doi.org/10.1016/j.landurbplan.2014.01.017 https://doi.org/10.1016/j.landurbplan.2012.06.003 https://doi.org/10.1016/j.landurbplan.2012.06.003 https://doi.org/10.1016/j. ecolind.2020.107080 https://doi.org/10.1016/j.landurbplan.2017.04.002 https://doi.org/10.1016/j.landurbplan.2017.04.002 https://doi.org/10.3390/ijgi10090596 https://doi.org/10.3390/ijgi10090596 https://doi.org/10.1016/j.habitatint.2016.05.006 https://doi.org/10.1016/j.habitatint.2016.05.006 https://doi.org/10.1016/j.ufug.2020.126603 https://doi.org/10.1016/j.isprsjprs.2017.09.007 https://doi.org/10.1016/j.isprsjprs.2017.09.007 Kolcsár, R.A. et al. Hungarian Geographical Bulletin 71 (2022) (3) 249–269.270