GIS Multicriteria Decision Analysis in Selecting the Optimal Location

for Urban Green Space: A Case Study of Zadar City

Rina Milošević

, Silvija Šiljeg

and Fran Domazetović

Croatian Geographical Society Zadar, Zadar, Croatia

Department of Geography, University of Zadar, Zadar, Croatia

Keywords: Urban Green Areas, GIS-MCDA, Optimal Location, City of Zadar.

Abstract: The urbanization process has proceeded rapidly in recent decades, resulting in the rapid transformation of

natural surfaces into impervious ones which has numerous impacts on the environment and human health.

Urban green spaces are recognized as a critical spatial component for maintaining ecological balance and

improving human mental and physical health. Therefore, the rational and even distribution of green spaces in

the city is particularly important, as they represent the most accessible natural environment for city dwellers.

The main objective of this study is to propose criteria and create a UGS suitability model (UGSSM) for the

urban area of Zadar. The model is generated by applying the GIS multicriteria decision analysis (MCDA) and

analytical hierarchical process (AHP). The model resulted in 580 ha of very high suitable (VHS) zones, mostly

located in the northwestern part of the city. However, only 0.05% (N=38) of VHS zones are consolidated

areas larger than 2 ha. Among VHS consolidated areas (>2 ha), the optimal one is depicted based on ownership

verification. This framework can be applied to other small cities with some minor modifications. For future

research, we suggest including residents with physical disabilities in the selection and landscaping of the

location.

1 INTRODUCTION

The process of urbanization has progressed rapidly in

recent decades and by United Nations projections will

inevitably continue to increase (UN, 2018). Urban

areas are already home to approximately half of the

world's population and most industrial activities

(Semeraro, 2021, The World Bank, 2020). Life in

highly urbanized areas affects residents’ living

rhythm, which has become faster and more stressful.

Additionally, urbanization affects the environment in

various ways transforming natural surfaces into

impervious such as roads, roofs, or parking lots. Some

direct effects are the occurrence of urban heat islands

(Song et al. 2015, Xu et al. 2022), pluvial floods, and

biodiversity loss (McDonald et al. 2013, Song et al.

2015).

To mitigate these negative effects the importance

of the natural environment is emphasized due to its

various benefits. Urban green spaces (UGS) are

considered one of the main components of urban

environments (Gupta et al. 2012) in the context of

recreation, social contributions (Nath et al., 2018),

health (Jennings & Bamkole, 2019), and

environmental outcomes (Hunter et al. 2019). It

maintains the urban ecological balance by affecting

the urban microclimate, purifying the air, and

reducing the risk of heat islands, soil erosion, and

pluvial or flash flooding (Hunter et al. 2019). UGS

positively impacts city residents' mental and physical

health by reducing stress, providing social contact,

enabling physical activity, and reducing exposure to

pollutants, noise, and excessive heat (Jennings &

Bamkole, 2019). They are also considered one of the

indicators of quality of life and housing in cities

(Šiljeg et al. 2018).

There are various definitions of UGS. According

to the Urban Green Belt project, UGS is defined as

any public or private open space covered with

vegetation, directly or indirectly accessible to users

(Šiljeg et al. 2018). The World Health Organization

defines it as any urban space that is covered with

vegetation and is crucial in promoting healthy living

conditions for all urban residents (WHO, 2017).

According to the ANGSt (Accessible Natural

Green Space standard) developed by Natural

England, green spaces are places where human

activities are not intense and natural processes prevail

Miloševi

c, R., Šiljeg, S. and Domazetovi

c, F.

GIS Multicriteria Decision Analysis in Selecting the Optimal Location for Urban Green Space: A Case Study of Zadar City.

DOI: 10.5220/0012031400003473

In Proceedings of the 9th International Conference on Geographical Information Systems Theory, Applications and Management (GISTAM 2023), pages 237-243

ISBN: 978-989-758-649-1; ISSN: 2184-500X

 2023 by SCITEPRESS – Science and Technology Publications, Lda. Under CC license (CC BY-NC-ND 4.0)

237

(English Nature, 2003, Šiljeg et al. 2020). This

standard emphasizes that green space is available

when it can be used without fees and time constraints.

In addition, the importance of size (area) and distance

of UGS are emphasized. The upgrade of this

methodology (ANGSt Plus) includes the concept of

connectivity, defined as the physical possibility of

access to green spaces. It also states the importance

of an even distribution of green spaces in urban areas

with respect to the socioeconomic characteristics of

the area.

The mentioned definitions are not encompassing

the importance of arrangement and mantainment the

UGS. However, to use a particular UGS for rest and

recreation some basic elements of accessibility must

be achieved. These include basic infrastructure such

as paths and benches, that do not significantly alter

the natural environment. This infrastructure is

especially important for people with physical

disabilities who often do not use the benefits of UGS

due to its lack of adaptation. The rationality and

equity of UGS distribution are particularly important

nowadays as the UGS is the most accessible natural

environment for city residents (Semeraro, 2021).

However, UGS management can be very challenging,

especially in densely populated city areas (Haaland et

al. 2015), and is often affected by the specific

characteristics of each urban area (Linh et al. 2022).

Some of the studies of UGS are focused on the

ecological suitability of the location (Li et al., 2018;

Linh et al. 2022). For the big cities, several general

criteria are typically used. However, for smaller urban

areas, the selected criteria are more specific to their

locations, involving their individual distances and

infrastructure (WHO, 2017; Linh et al. 2022).

Therefore, the main goal of this study is to

determine the optimal location for UGS landscaping

in the city of Zadar. The mentioned study case is

depicted because Šiljeg et al. 2018 performed a UGS

accessibility analysis and pointed out the lack of

landscaped UGS in the city. The multiple-criteria

decision analysis (MCDA), and AHP process were

used to determine additional potential locations for

UGSs arrangement.

2 STUDY AREA

Zadar (25 km²) is the administrative center of Zadar

county (Figure 1). In the last decades, Zadar is

characterized by intense urban sprawl followed by an

increase in urban population due to dynamic and

strong development as a regional center (Magaš,

1991, Magaš, 2011). At the latest census (2021), the

total population was 75,082 (DZS, 2021). Recent

economic and demographic trends are also reflected

in urban physiognomy. For example, the old city

center (Peninsula) is predominantly the central

business district, while peripheral parts (Novi

Bokanjac, Dračevac, Smiljevac) are now having a

residential role (Graovac, 2004). Rapid urbanization

also resulted in an increase in impervious surfaces

and a lack of UGS (Šiljeg et al. 2018).

Figure 1: Study area.

3 MATERIALS AND METHODS

The main goal of this study is to suggest criteria and

create the UGS suitability model (UGSSM) for the

urban area of Zadar. To create the model GIS-MCDA

and Analytical hierarchical process (AHP) were

used. Since the Zadar is a smaller urban area, the

selected criteria are specific to the research area

involving individual distances and infrastructure.

GISTAM 2023 - 9th International Conference on Geographical Information Systems Theory, Applications and Management

238

3.1 GIS - MCDA

The GIS-MCDA is one of the most popular

procedures used to estimate the suitability of land for

various purposes (Modica et al. 2014). The procedure

consists of 1) identification of the problem and

defining the goal, 2) criteria selection, 3) criteria

standardization, 4) calculation of weight coefficients,

5) criteria aggregation, and 6) model validation

(Domazetović et al. 2019) (Figure 2).

Figure 2: Methodology workflow.

To create UGSSM, criteria were derived in the

ArcGIS environment using Network Analyst (Service

area), Spatial Analyst (Slope, Point Density, Raster

Calculator), and Analysis tool (Proximity - Buffer).

Input data include WorldView 2 (WV-2)

multispectral (MS) imagery and OpenStreetMap

(OSM) data. The land cover model (LULC) is

generated from the WV-2 MS imagery, using the

Geographic Object-Based Image Analysis

(GEOBIA) and Support Vector Machine (SVM)

algorithm.

To be comparable, all derived criteria (except

Boolean) were standardized to scale from 1 to 5 (1 –

very low suitability, 5 – very high suitability) using

the Jenks method. Boolean criteria were standardized

to a binary scale (0-false and 1-true). An AHP was

used to rank criteria according to their level of

suitability and to calculate the weighting coefficient

for each (W

). For the matrix validation consistency

ratio was calculated (CR) (Figure 3).

Figure 3: Pairwaise matrix (AHP).

The following formula is used to aggregate the

criteria (Eastman 1999):

P = ∑ 𝑤𝑖 𝑋𝑖 * ∏ 𝐶𝑗

where: P = suitability 𝐶𝑗 = restriction 𝑤𝑖 =

weighted coefficient ∑ =sum of weighted criteria; ∏

= sum of restrictions (1 – suitable, 0 – unsuitable) 𝑋𝑖

= criteria value.

3.2 Selection of Optimal Location

The generated GIS-MCDA suggests the most suitable

location for UGS based on the selected criteria.

However, verifying public availability, i.e.,

ownership of potential UGS is necessary. Verification

of ownership is a complex process due to the large

number of cadastral parcels in the Zadar. Therefore,

ownership is verified only for the very high UGS

suitability classes. The ownership verification is

performed by overlaying the GIS-MCDA model with

the official cadastral parcel map prepared by the State

Geodetic Administration (DGU). In addition, the

UGSs with a larger area have priority in the selection

of the optimal site.

4 RESULTS AND DISCUSSION

4.1 Criteria Selection

To select the optimal location for UGS seven criteria

were used: land cover model (LULC), slope,

residential object density, accessibility, road distance,

NDVI, and constraints (Boolean).

GIS Multicriteria Decision Analysis in Selecting the Optimal Location for Urban Green Space: A Case Study of Zadar City

239

4.1.1 LULC and Boolean

Land cover has an important role in the selection of

locations for UGSs as it determines the feasibility of

UGSs (Linh et al. 2022). Therefore, it is ranked as the

most important criterion in selecting the location for

UGS in the urban area of Zadar.

The land cover model consists of a total of eight

classes (Figure 4). The existing UGS such as lawns

and forest vegetation were rated as the most suitable

classes for arrangement. The Boolean criterium is

based on LULC and all build-up areas (impervious

and objects), bare land (soil), and water surfaces are

classified as constraints.

Figure 4: Land cover model (LULC).

4.1.2 Slope

Generally, low-incline areas are considered more

appropriate for the UGS as it will determine the soil

characteristics and rate of erosion (Sharma et al.

2022). Accordingly, gentle slopes are classified as

the most suitable for UGS landscaping. Although

the most of urban area of Zadar is characterized by

a gentle slope, the most suitable zones are located in

the north-western part of the city (Figure 3). Ovo je

dio grada

4.1.3 Residential Objects Density

The residential objects’ density represents the

spatial distribution of the population. To make UGS

accessible to as many people as possible, areas with

a higher density of residential buildings were

assessed as more suitable. The highest density of

residential buildings is in the central part of the city

(Figure XX), while the density decreases towards

the outskirts where most of the arable agricultural

areas are located.

Figure 5: Slope.

Figure 6: Residential objects density.

4.1.4 NDVI

NDVI represents the vegetation’s health and helps

distinguish areas with poorer vegetation health from

areas with healthier and dense vegetation. The higher

values of NDVI indicate more suitable locations for

UGS (Linh et al. 2022). In the area with an NDVI

value lower than 0.1 UGS is deemed inappropriate.

NDVI was extracted from the multispectral WV2

satellite image using the following formula:

𝑁𝐷𝑉𝐼 





(1)

GISTAM 2023 - 9th International Conference on Geographical Information Systems Theory, Applications and Management

240

Figure 7: NDVI.

4.1.5 Accessibility

The accessibility refers to the amount of time required

to get from the residential buildings to the first UGS.

According to European Environment Agency (EEA),

an individual should have access to a green space

within 15 min (walking distance) of their residence

(David et al., 1995). However, this distance differs

depending on the region (Lieh, 2022). Using the

Network Analyst Service area tool five zones of

accessibility were created for Zadar urban area: less

than 5 minutes, 5 - 10, 10 - 15, 15- 20, and more than

20 minutes (walking) from residential buildings. Due

to the dense road network most of the urban area of

Zadar is accessible within 5 minutes (Figure 8).

Figure 8: Accessibility from residential units (minutes).

4.1.6 Distance to Main Roads

The distance from roads is important regarding

people’s ability to easily access UGS areas (Linh et

al. 2022). In this study, main roads include all road

infrastructure together with pedestrian and bicycle

routes. Macadam roads on the outskirts of the city

were excluded from the analysis due to inaccessibility

and poor maintenance. Using the Multiple buffer tool

five classes of road distance are derived: 0-100 m,

100-250 m, 250-500 m, 500-1000 m, and more than

1000 m. Most of the city area is within less than 100

meters distance from roads (Figure 9).

Figure 9: Road distance.

4.1.7 Urban Green Spaces Suitability Model

GIS-MCDA resulted in a total of 64,128 polygons

representing the most suitable UGS (5- very high

suitability (VHS)) (Figure 7). The total area of VHS

is 558 ha.

VHS land classes are mostly located in the

western part of the city, while the very low suitable

classes are predominately located in the old city

centrum. High-suitable classes are located in the

eastern part of the city, near the industrial zone

“Gaženica”. Arranging UGS in this part of the city

would maintain the ecological balance. However,

UGS in this part of the city is not accessible from

many residential objects within 5 or 10 minutes.

4.1.8 Selection of Optimal Location

To select the optimal location two additional criteria

were included: size and ownership. From the total

number of generated VHS classes, only polygons

(N=38) with an area higher than 2 hectares were

GIS Multicriteria Decision Analysis in Selecting the Optimal Location for Urban Green Space: A Case Study of Zadar City

241

Figure 10: GIS-MCDA suitability model.

extracted (Figure 8). This criterium is included

because of ANGst’s suggestion: each resident should

have at least one UGS of more than 2 ha in size at a

maximum distance of 300 m. Based on the ownership

verification, it appears that almost all very high

suitability land classes are privately owned and

cannot be used by the public. Only one UGS is

property of Zadar city and can potentially be

landscaped (Figure 11).

Figure 11: Extracted UGS (> 2 ha and very high suitable)

and selected optimal location for potential UGS in Zadar.

5 CONCLUSIONS

This paper represents an attempt to improve UGS

accessibility in small urban areas by suggesting

criteria and conducting the GIS-MCDA process. The

research has shown that GIS-MCDA is an efficient

technique that can be used in urban green

infrastructure planning and related fields. Created

suitability model resulted in several zones of very

high suitability. Among these zones, the optimal one

is depicted based on ownership verification. For

future research, we suggest including residents in the

selection of the location. To gather information about

their needs and preferences it is recommendable to

conduct a public perception survey. In addition, it is

particularly important to include people with

disabilities in research to recognize the measures that

can be taken to improve accessibility. With some

minor modifications, this framework can be applied

to other small cities. The results of this research could

be useful to decision-makers in developing land use

plans.

ACKNOWLEDGEMENTS

This paper is conducted as part of DORAS project

UP.04.2.1.11.0172. Non-refundable funds were

awarded by European Social Fund (85%), as well as

the State budget of the Republic of Croatia (15%). For

more info on funds, visit www.strukturnifondovi.hr

and www.esf.hr

REFERENCES

Bell, D., & Jayne, M. (2006). Conceptualizing small cities.

In Small Cities (pp. 1-18). Routledge.

David, S. Europe’s Environment—The Dobris Assessment;

Philippe, B., Ed.; European Communities: Brussels,

Belgium, 1995; ISBN 92-826-5409-5.

David, S. Europe’s Environment—The Dobris Assessment;

Philippe, B., Ed.; European Communities: Brussels,

Belgium, 1995; ISBN 92-826-5409-5.

Domazetović, F., Šiljeg, A., Lončar, N., & Marić, I. (2019).

Development of automated multicriteria GIS analysis

of gully erosion susceptibility. Applied geography, 112,

102083.

Eastman, J. R. (1999). Multi-criteria evaluation and GIS.

Geographical information systems, 1(1), 493-502.

English Nature (2003). Providing accessible natural

greenspaces in towns and cities: a practical guide to

assessing the resources and implementing local

standards for provision, UK.

GISTAM 2023 - 9th International Conference on Geographical Information Systems Theory, Applications and Management

242

Graovac, V. (2004). Populacijski razvoj Zadra. Geoadria,

9(1), 51-72.

Gupta, K., Kumar, P., Pathan, S. K., & Sharma, K. P. (2012).

Urban Neighborhood Green Index–A measure of green

spaces in urban areas. Landscape and urban planning,

105(3), 325-335.

Haaland, C.; van den Bosch, C.K. Challenges and strategies

for urban green-space planning in cities undergoing

densification: A review. Urban For. Urban Green. 2015,

14, 760–771.

Handy, S., Heinen, E., & Krizek, K. (2012). Cycling in

small cities. City cycling, 257-286.

Hunter, R. F., Cleland, C., Cleary, A., Droomers, M.,

Wheeler, B. W., Sinnett, D., ... & Braubach, M. (2019).

Environmental, health, wellbeing, social and equity

effects of urban green space interventions: A meta-

narrative evidence synthesis. Environment international,

130, 104923.

Jennings, V., & Bamkole, O. (2019). The relationship

between social cohesion and urban green space: An

avenue for health promotion. International journal of

environmental research and public health, 16(3), 452.

Li, Z., Fan, Z., & Shen, S. (2018). Urban green space

suitability evaluation based on the AHP-CV combined

weight method: A case study of Fuping county, China.

Sustainability, 10(8), 2656.

Linh, N. H. K., Tung, P. G., Chuong, H. V., Ngoc, N. B., &

Phuong, T. T. (2022). The Application of Geographical

Information Systems and the Analytic Hierarchy

Process in Selecting Sustainable Areas for Urban Green

Spaces: A Case Study in Hue City, Vietnam. Climate,

10(6), 82.

Magaš, D. (1991). Neke promeje u gradskim naseljima

zadarske regije u razdoblju 1857.-1971. godine. Radovi

Zavoda za povijesne znanosti HAZU u Zadru, (33),

239-258.

Magaš, D. (2011). Koncept teritorijalnog ustroja Jadranske

Hrvatske. Geoadria, 16(2), 211-236.

McDonald, R. I., Marcotullio, P. J., & Güneralp, B. (2013).

Urbanization and global trends in biodiversity and

ecosystem services. Urbanization, biodiversity and

ecosystem services: challenges and opportunities: a

global assessment, 31-52.

Modica, G., Laudari, L., Barreca, F., & Fichera, C. R.

(2014). A GIS-MCDA based model for the suitability

evaluation of traditional grape varieties: the case-study

of ‘Mantonico'Grape (Calabria, Italy). International

Journal of Agricultural and Environmental Information

Systems (IJAEIS), 5(3), 1-16.

Nath, T. K., Han, S. S. Z., & Lechner, A. M. (2018). Urban

green space and well-being in Kuala Lumpur, Malaysia.

Urban Forestry & Urban Greening, 36, 34-41.

Semeraro, T., Scarano, A., Buccolieri, R., Santino, A., &

Aarrevaara, E. (2021). Planning of urban green spaces:

An ecological perspective on human benefits. Land,

10(2), 105.

Sharma, Richa, Pradhan, Lolita, Kumari, Maya and

Bhattacharya, Prodyut. "Urban Green Space Planning

and Development in Urban Cities Using Geospatial

Technology: A Case Study of Noida" Journal of

Landscape Ecology, vol.15, no.1, 2022, pp.27-46.

https://doi.org/10.2478/jlecol-2022-0002

Šiljeg, S., Marić, I., Nikolić, G., & Šiljeg, A. (2018).

Accessibility analysis of urban green spaces in the

settlement of Zadar in Croatia. Šumarski list, 142(9-10),

487-496.

Šiljeg, S., Milosevic, R., & Vilic, E. (2020). Multiscale GIS

based Analysis of Urban Green Spaces (UGS)

Accessibility: Case Study of Sisak (Croatia). In

GISTAM (pp. 240-245).

Song, Y., Li, F., Wang, X., Xu, C., Zhang, J., Liu, X., &

Zhang, H. (2015). The effects of urban impervious

surfaces on eco-physiological characteristics of Ginkgo

biloba: A case study from Beijing, China. Urban

Forestry & Urban Greening, 14(4), 1102-1109.

The World Bank. Urban Population. Available online:

https://data.worldbank.org/indicator/SP.URB.TOTL.I

N.ZS (accessed on 19 July 2020).

United Nations (2018). World population to reach 9.7

billion by 2050 - UN, retrieved from (URL 4

https://www.dw.com/en/world-population-to-reach 97-

billion-by-2050-un/a-49241728) Accessed at 02

January 2020.

World Health Organization. Urban Green Spaces: A Brief

for Action; World Health Organization: Copenhagen,

Denmark, 2017

Xu, C., Chen, G., Huang, Q., Su, M., Rong, Q., Yue, W., &

Haase, D. (2022). Can improving the spatial equity of

urban green space mitigate the effect of urban heat

islands? An empirical study. Science of The Total

Environment, 841, 156687.

GIS Multicriteria Decision Analysis in Selecting the Optimal Location for Urban Green Space: A Case Study of Zadar City

243