Kompaktowość polskich terenów zurbanizowanych – metody badań i analiza z użyciem bazy danych CLC

Ewa Szymczyk, Mateusz Bukowski

doi:10.37190/arc250112

Streszczenie

W artykule autorzy przedstawili dynamikę kompaktowości (zwartości) zabudowy miejskiej w Polsce, analizując metody oceny tego zjawiska oraz zmiany zachodzące w czasie. Trzy główne pytania badawcze skoncentrowali wokół metod analizy zwartości, związku pomiędzy wielkością gmin miejskich a zwartością oraz zmian czasowych. Badaniem objęto wszystkie polskie gminy i zanalizowano wybrane miary zwartości w latach 2006, 2012 i 2018. Poprzez ocenę różnych metod analitycznych oraz wykorzystanie danych o użytkowaniu terenu z bazy CORINE Land Cover autorzy zaproponowali nową metodologię monitorowania wskaźników zwartości miast, wspierając tym samym podejmowanie decyzji planistycznych opartych na danych, dążąc do zrównoważonego planowania przestrzennego.

Wyniki ujawniają wzorce i korelacje pomiędzy wielkością miast a zwartością w czasie, wskazując na ogólny spadek zwartości zabudowy w badanym okresie (2006–2018). Wykazano istnienie logarytmicznego związku między wielkością populacji, indeksem zwartości i gęstością zaludnienia obszarów miejskich. Mniejsze miasta cechują się większą zwartością kształtu, ale niższą gęstością, natomiast większe miasta są gęściej zaludnione, lecz jednocześnie bardziej rozproszone. Praca wnosi wkład do stanu wiedzy, wprowadzając i testując metodę oceny zwartości zabudowy miejskiej w różnych typach gmin. Ponieważ dane CLC obejmują cały obszar Unii Europejskiej, metoda ta może być replikowana w każdym państwie członkowskim, umożliwiając dalsze badania porównawcze.

Podgląd pełnego artykułu jest możliwy wyłącznie na większych ekranach.

2025

1(81)

Ewa Szymczyk*, Mateusz Bukowski**

Compactness of Polish urban areas

– methodologies and analysis based on CLC dataset

DOI: 10.37190/arc250112

Published in open access. CC BY NC ND license

Abstract

This research delves into the dynamics of urban compactness in Poland, exploring methodologies for analysis and assessing changes over time.

Three key research questions guiding the study revolve around methods for compactness analysis, the relationship between the size of urban munic-

ipalities and compactness, and temporal changes.

The research considers all Polish municipalities and analyses selected compactness measures in 2006, 2012 and 2018. By evaluating various ana-

lytical approaches and utilising land use data from CLC (CORINE Land Cover), the study provides a new methodology to monitor city compactness

measures, thus contributing to evidence-based decision-making for sustainable urban planning.

The outcomes unveil patterns and correlations across city sizes and temporal trends, showing the decrease in compactness over time. It reveals

a logarithmic relation between population size, compactness index, and urban population density. While smaller cities tend to be more compact in

shape, they are less dense. Conversely, larger cities tend to be more densely populated but also more spatially dispersed. This study contributes to

the state of knowledge by introducing and testing a method to assess urban compactness across urban areas. Since the CLC datasets cover the entire

European Union, this method is replicable in every member state, allowing for further comparative studies.

Key words: Poland, compactness, urban population density, CLC, urban form measures

Introduction

There has been a growing interest in urban patterns in

European and North American literature, specically fo

cusing on quantitative methods (Reis, Silva and Pinho

2016). The emphasis on sustainable development models,

the advancement of GIS tools and information technol-

ogy, and the increased availability and quality of spatial

data have led to a rapid acceleration of data-driven spatial

research across various domains. Geographers, environ-

mental scientists, economists, and policymakers have all

analysed the relationships with spatial patterns. Spatial in-

dicators are crucial in examining the connections between

the built environment and dierent urban processes. It is

similar in urban science, where spatial measures are essen-

tial for monitoring the development of spatial organisation

within cities over time, allowing for analysing and com-

paring spatial-temporal patterns. While numerous methods

are avail able for such analyses, the choice of method is ty-

pi cally inuenced by its relevance to the specic question.

Urban form measures have become a central focus in the

study of urban compactness, an urban form characteristic

strongly linked to sustainable development (OECD 2012)

and seen as the opposite of urban sprawl (Tsai 2005). At the

same time, urban sprawl is known to have high social costs

in urban planning (Squires 2002; Schwarz, Haase and Sep-

pelt 2010; Siedentorp, Fina 2010; UN-

HABITAT 2022);

recent research suggests that urban forms signicantly im-

pact commuting patterns (Song et al. 2017).

According to

Schiller and Kenworthy (2018), this development pattern

also leads to increased costs, adverse eects on the city

centre, higher energy and fuel consumption, and detri-

mental eects on household budgets and the environment.

In a study on the relationship between urban sprawl and

* ORCID: 0000-0003-3698-3478. Faculty of Architecture and

Urban Planning, Cracow University of Technology, Poland, e-mail: ewa.

szymczyk@pk.edu.pl

** ORCID: 0009-0001-0285-8241. Independent researcher, Swi-

tzer land.

130 Ewa Szymczyk, Mateusz Bukowski

economic performance in growing Polish cities, Lityński

(2021) found that stronger local economies are associated

with less urban sprawl. Conversely, high urban compact-

ness and high urban density are believed to benet public

transit systems (OECD 2012; Schiller, Kenworthy 2018).

The concept of the compact city is one of the most wide-

ly discussed in contemporary urban policy (OECD 2012).

The term compact city was rst used by Dantzig and Saaty

(1973), who were principally interested in a more ecient

use of urban resources. It is recommended for urban devel-

opment at all levels of planning guidelines, from global or-

ganisations like the UN-HABITAT to the European Union

and in Polish national planning documents (Uchwała

2015). Compactness can be measured in various ways but

generally relates to urban expansion and density patterns.

The OECD denes the characteristics of the compact city

as […] dense and proximate development patterns […]

urban areas linked by public transport systems […] ac-

cessibility to local services and jobs (OECD 2012, 15).

Understanding urban compactness is crucial for address-

ing urban sprawl and promoting and eectively planning

sustainable urban development globally and locally. Since

this concept in Poland was mainly analysed from the per-

spective of the opposite phenomenon of urban sprawl and

never from a nationwide scale, this study lls this gap. It

aims to achieve the methodological and empirical objec-

tives of assessing urban compactness in Poland. This aim

is achieved by addressing three key research questions:

1. What are the methods for analysing compactness?

2. What is the relationship between city size and city

compactness in Poland?

3. How does compactness change over time in Polish

cities?

The research assesses various methods for analysing

urban compactness, including morphological indices and

computational models, and their relevance in the context

of Poland. It suggests a methodology based on the CLC

(CORINE Land Cover), a crucial data source on Europe’s

land use and landscape dynamics. The study applies two

methods to assess the compactness of all municipalities in

Poland. It uses historical CLC data to explore the evolu-

tion of urban compactness in all Polish urban areas (in all

municipalities) and separately only for urban areas in cities

(urban and urban-rural municipalities) and its relationship

to the municipality population size.

State of research

Ahlfeldt and Pietrostefani (2017) reviewed the theoret-

ical literature and identied three main characteristics of

compact cities: economic density, morphological density,

and mixed land use. Each of these characteristics can be

measured in various ways. Economic density, the number

of people or jobs within a given area, is typically mea-

sured using population or employment density (Ahlfeldt,

Pietrostefani 2017). Morphological density focuses on the

built environment and includes measures such as

urban/

rural boundaries, street connectivity, and building foot-

print-to-parcel ratio. Mixed land use captures the co-loca-

tion of dierent functions, such as residential, employment,

and retail. It can be measured in two dimensions (Ahlfeldt,

Pietrostefani 2017) or vertically (in three

dimensions) (Bur-

ton 2002). When considering the concept of a compact

city, it is essential to select measures and methods based

on their relevance to the specic question being addressed,

available data and the scale of the analysis (Ottensmann

2021).

A study on spatial metrics conducted by Reis et al.

(2015) provides an extensive overview of all measures

and methods to study patterns of urban growth and shrink-

age. Authors subdivided patterns of growth into four main

groups:

1) expansion,

2) urban sprawl,

3) polycentrism,

4) densication/coalescence.

According to their review, urban sprawl is the most

studied pattern in Planning and Geography. While the ex-

act denition is ambiguous, it is commonly understood as

characterised by scatter/fragmentation, low density (both

population and building), single-use and poor accessibility

(Reis et al. 2015). Compactness, seen as the opposite of

urban sprawl, may be measured similarly.

The most commonly used measure of compactness is

related to density (Burton 2002). Most known research

uses population density (for example, Newman and Ken-

worthy 1989). When it comes to spatial measures, the com-

pact city concept typically concerns the two-dimensional

expansion pattern of an urbanised area, which is consid-

ered more compact if the pattern is more clustered towards

a centre and with less sprawl, leap-frogging or branching

(Mubareka et al. 2011). Referring to Reis et al. (2015)

classication of metrics, the analysis can use one of three

metrics groups depending on the area of knowledge and

methodological approach to urban form. These groups are

Landscape metrics, Geospatial metrics and Spatial metrics.

Categories organise each of these groups and consist of

numerous metrics to investigate them. Figure 1 shows how

the selected shape metrics t into the classication.

Ottensmann (2021) conducted a study on the urban

shape of 59 large cities in the US, evaluating measures

based on specic criteria. The author identied three cru-

cial aspects for determining the most suitable shape indi-

ces: the ability of metrics to measure the relation to the city

centre or CBD (Central Business District in the US), the

method of measuring holes and discontinuous areas, and

the applicability to given urban area data. The study re-

viewed shape measures that assess the compactness of ur-

ban areas, assuming that the circle is the most compact 2D

shape. In his research on large US cities, Ottensmann used

the Proximity Index (belonging to the “Distances to Areas

in the Shape” category). This landscape metric measures

shape irregularity by calculating the distance from all loca-

tions within the urban area to the CBD. The author points

out that this method is suitable when only one centre can

be identied but is not useful for polycentric urban areas.

Similarly, Burton (2002) studied English compactness,

selecting 25 towns, medium-sized, and big cities. How-

ever, Burton developed a multi-criteria indicator method.

Indicators were organised into six groups: compactness,

Compactness of Polish urban areas – methodologies and analysis based on CLC dataset 131

density, mixeduse, intensication, population intensi

ca tion, and builtform intensication, each with four vari-

ables. While the measures of compactness objectively re -

presented the phenomenon, the author concluded that they

may not always accurately represent subjective compact-

ness. This supports the idea that compactness is a socio-

cultural construct rather than a purely objective one.

Thinh et al. (2002) research on the compactness of 116

German cities was driven by the question of how compact

a sustainable city is. The authors used data from the CLC

and the Digital Landscape Model (DLM) of Germany to

establish and validate the degree of surface sealing (in

percentage). They established a GIS raster analysis using

a square raster (500/500 m grid) and the gravitational ap-

proach (Thinh et al. 2002). Their research presented a con-

nection between various socio-environmental indicators

and the compactness of city form.

In the literature about Polish cities, there is a signi-

cant focus on urban sprawl. Urban compactness is primar-

ily examined in terms of urban density. Śleszynski (2014)

conducted the most extensive study on urban population

density in 147 Polish cities. His analysis utilised data from

the 2002 census, which divided cities with over 30,000 in-

habitants into statistical districts, enabling more in-depth

studies on population distribution within cities. The key

ndings are related to the distribution of density from the

city centre, revealing a strong correlation between popu-

lation density in the 0–2 km concentration zone from the

city centre and the demographic sizes of towns or cities.

Smaller cities and towns generally have less dense centres

compared to larger ones. Lityński (2021) conducted a ma-

jor study on urban sprawl in relation to economic perfor-

mance. The Author analysed 4 big Polish cities and their

neighbouring municipalities. The method used to assess

urban sprawl is based on the following indicators: density,

continuity, concentration, clustering and centrality. Higher

ratios indicate less urban sprawl. Density, for example, is

the number of housing units per hectare of urbanised land

Author calls it Developable land (DL) which is an area that is the

dierence between the total municipal area and the sum of land covered

by water, forest, recreation areas, roads and land reserved for ecological

purposes (Lityński 2021).

and centrality is the degree to which buildings are located

in relation to the city centre. It calculates the distance from

the village centre to the city centre. The measure is the in-

verse of the average sum of these distances weighted by

the number of housing units in the village.

The methods listed above answer the rst research

question by demonstrating the various ways compactness

can be measured, depending on the specic focus of the

research. The next sections of this study add to the method-

ologies for measuring the compact city by oering a tech-

nique that utilises open-source spatial data (CLC) and sta-

tistical population data.

Data and methods

This study adopts a quantitative approach to examine

Poland’s urban municipalities from the perspective of

12 years. The selection of subjects is based on Poland’s

administrative system. On the highest/regional level,

there is a voivodeship (województwo), which consists

of counties (powiat); counties are divided into individu-

al local-authority municipalities (gmina); a municipality

contains either an individual city, which by GUS (Główny

Urząd Sta tys tyczny, Eng. Statistics Poland) classication:

level 6, kind 1), only villages – rural area (GUS classi-

cation: level 6, kind 2), or a mix of a town and villages

– urbanrural area (GUS classication: level 6, kind 3)

(GUS BDL 2023). The selected method uses the small-

est local government unit, i.e., the municipality (gmina).

In 2021, Poland had 2,477 municipalities: 302 urban, 662

urban-rural and 1,513 rural. Therefore, the analysis is di-

vided into three administrative categories:

– an urban area in the urban municipality (GUS level 6,

kind 1),

– an urban area in the urban-rural municipality (GUS

level 6, kind 3), including a city and small settlements,

– an urban area in a rural municipality (GUS level 6,

kind 2), including towns and small settlements.

The varying spatial units over dierent years in Poland

were carefully considered to maintain data consistency.

Additionally, to see the possible linkage with the size of

urban areas, this study relied on a simplied division based

on Statistics Poland’s categorisation:

Fig. 1. Landscape metrics groups

and shape metrics categories

(elaborated by E. Szymczyk,

based on Reis et al. 2015;

Ottensmann 2021)

Il. 1. Grupy metryk przestrzennych

i kategorie metryk kształtu

(oprac. E. Szymczyk,

na podstawie Reis et al. 2015

oraz Ottensmann 2021)

132 Ewa Szymczyk, Mateusz Bukowski

– below 20,000 as a small city,

– 20,000–100,000 as a medium city and

– above 100,000 a big city.

The population was categorised based on the year of the

spatial analysis data.

The timeframe of our study was chosen for two reasons:

rst, because the most recent data available at the time was

provided for 2018, and second, to see urban areas for an

extended amount of time after the European Union acces-

sion (in 2004). The data for urban spatial metrics is based

on land use data from the CLC for the area of Poland. The

CLC data is a standardised methodology for producing

continent-scale land cover, biotope, and air quality maps,

including 44 land-use classes. Since the product is updated

every six years, with the most recent update being in 2018,

the three data sets were selected: CLC2006, CLC2012, and

CLC2018.

To dene the shape of the urban area, the 44 classes of

land used by the CLC were considered. Urban areas were

broad ly dened as “articial surfaces” (CLC), which in-

clude infra struc ture and green urban areas, represented by

11 classes (Table 1). This method is similar to the one fol-

lowed by the GUGiK (Główny Urząd Geodezji i Karto gra

i, Eng. Geodesy and Cartography Poland) (GUGiK 2024).

While most of the population lives in the rst two “urban

areas” classes

(111, 112),

the remaining nine classes repre-

sent urbanised areas, which are an integral part of the city.

The selected classes were cut to the municipality’s

boundaries and merged into one shape consisting of one

or more polygons (Fig. 2). Each shape received a unique

identication number of a spatial unit, called TERYT

CLC Code Name

111 Continuous urban fabric

112 Discontinuous urban fabric

121 Industrial or commercial units

122

Road and rail networks

and associated land

123 Port areas

124 Airports

131 Mineral extraction sites

132 Dump sites

133 Construction sites

141 Green urban areas

142 Sport and leisure facilities

Table 1. CLC Codes indicate classes taken under consideration

and their respective colours

(source: CLC 2023)

Tabela 1. Kody CLC wskazujące klasy uwzględnione w analizie

oraz odpowiadające im kolory

(źródło: CLC 2023)

Fig. 2. Diagram presenting a stepbystep spatial analysis method, defining compactness index and urban population density based on Jasło

(elaborated by E. Szymczyk)

Il. 2. Schemat przedstawiający metodę analizy przestrzennej, definiującą indeks zwartości (Ci) oraz gęstość zaludnienia obszarów miejskich (Ud)

na przykładzie Jasła (oprac. E. Szymczyk)

Compactness of Polish urban areas – methodologies and analysis based on CLC dataset 133

(Krajowy rejestr urzędowy podziału terytorialnego kraju,

eng. National Ocial Register of the Territorial Division

of the Country). Moreover, the development of the ana-

lytical method included some calibrations. Due to some

discrepancies in the spatial data (such as imprecise land

use demarcation), many cities had minor shapes left after

cutting to the boundaries. After analysing possible thresh-

olds to avoid distorting the calculation, a threshold of 2%

of the total shape area was set. All the shapes that were

below this value were deleted (step D).

The calculation consists of the following operations for

each selected year:

A. Isolation of selected 11 land use classes.

B. A cut of land use polygons with city administrative

boundaries.

C. Merge polygons into continuous shapes.

D. Delete polygons below a set threshold – for a given

shape.

E. Calculate the shape’s area and perimeter.

F. Calculate the compactness index.

G. Calculate the urban population density.

As presented in State of research section, there are mul-

tiple ways to measure urban area compactness. This study

uses the urban compactness (Ci) and urban density (Ud)

index with values in persons per square kilometre.

The compactness index quanties the compactness of

shapes’ irregularity. There are many measures of the com-

pactness index (Niemi et al. 1990; Altman 1998; Cham-

bers, Miller 2010). A few were identied by Barnes and

So lomon (2021) as the most commonly used. Among

these most common, the authors selected a single measure

named Schwartzberg (1966). The Schwartzberg compact-

ness index is determined by comparing the perimeter (Pd)

of an urban area (see Fig. 2) to the circumference of a cir-

cle that has the same area (Ad) as the urban area. It can be

expressed as follows:

Index ranges from 0 to 1, where 0 is the least compact

and 1 is the most compact. This index was calculated for

each medium-sized city in each CLC year.

The second measure considers the urban area (Ua)

and the municipality population (Pop) in the correspond-

ing year to calculate the urban population density. Com-

pared to the simplied population density found in GUS,

this method of selecting urban areas excludes agricul-

tural land, forests and water bodies, among others, giv-

ing a more accurate picture of where the city urbanised

boundaries. The urban population density (Ud) can be for-

mally written as:

Results

This section presents the analysis outcomes for all mu-

nicipalities (2477) and separately only for urban munici-

palities (964) in three periods based on the CLC datasets.

The outcomes are divided into three relations:

– trends in time,

– trends concerning city size,

– trends concerning the location.

Firstly, the compactness index (Ci) of urbanised areas

in Polish municipalities shows a decreasing trend over

time. This is true when urban areas in all municipalities are

considered (Fig. 3a) and when only the urban municipali-

ties (Kind 1, Kind 3 in GUS) are taken into consideration

(Fig. 3b). On average, urban municipalities were more

compact than rural ones. Table 2 shows detailed analysis

outcomes, revealing the decrease in the Ci of urban areas

in urban municipalities. The most compact Polish urban

municipalities in 2006 were Ząbki, Podkowa Leśna and

Wysokie Mazowieckie in 2012 and in 2018, it was Za-

widów, Wysokie Mazowieckie and Podkowa Leśna with

over 0,6 Ci. All the top-scoring cities are small (except for

Ząbki, a mediumsized city), and all are adjacent to a big

urban area.

Fig. 3. Compactness index trend for: a) all Polish municipalities, b) all urban municipalities (elaborated by M. Bukowski, E. Szymczyk)

Il. 3. Trend indeksu zwartości dla wszystkich polskich: a) gmin, b) gmin miejskich (oprac. M. Bukowski, E. Szymczyk)

a b

 =





2π

√





/π

 =





134 Ewa Szymczyk, Mateusz Bukowski

Secondly, the urban density of urbanised areas in Polish

municipalities over time was compared. Similar to com-

pactness, change over the years for all municipalities and

urban municipalities shows a trend of signicant decrease

in urban population densities (Figs. 4a, b). Interestingly,

the densest Polish urban municipality in 2006 – Sopot,

with over 5894 people/km

was decreasing its density to

reach 5413 people/km

in 2018. The densest urban munic-

ipalities in 2006 were Sopot, Świętochłowice and Chełm-

no. In 2012 and 2018, Wejherowo, Chełmno and Sopot

with over 5 thousand people/km

. For comparison, the

densest European city and one of the densest in the world

among mega cities – Paris, has an urban population density

of 20 641 people/km

. Table 3 shows detailed outcomes of

the analysis, revealing the decrease in the urban population

density in all urban municipalities.

Table 3. Outcomes of compactness index analysis of urban areas

in all urban municipalities (kind 1, kind 3)

(elaborated by E. Szymczyk)

Tabela 3. Wyniki analizy indeksu zwartości zabudowy

dla obszarów miejskich we wszystkich gminach miejskich

(rodzaj 1, 3 wg. GUS BDL) (oprac. E. Szymczyk)

Table 2. Outcomes of urban population density analysis

for all urban municipalities (elaborated by E. Szymczyk)

Tabela 2. Wyniki analizy gęstości zaludnienia obszarów miejskich

dla wszystkich gmin miejskich (oprac. E. Szymczyk)

Years 2006 2012 2018

Number of

observations

964 964 964

Median 0.1962 0.1912 0.1895

Mean 0.2239 0.2177 0.2158

Lowest outcome 0.0864 0.0929 0.0917

Highest outcome 0.4478 0.4251 0.4195

Fig. 4. Urban population density trend for: a) all municipalities, b) all urban municipalities

(elaborated by E. Szymczyk)

Il. 4. Trend gęstości zaludnienia dla wszystkich polskich: a) gmin, b) gmin miejskich

(oprac. M. Bukowski, E. Szymczyk)

Years 2006 2012 2018

Number of

observations

964 964 964

Median

[people/km

]

2183.1 1970.24 1945.08

Mean

[people/km

]

2487.4 2232.33 2173.55

Lowest outcome

[people/km

]

451.86 388.22 464.36

Highest outcome

[people/km

]

5353.09 4730.06 4611.62

The outcomes of compactness analysis for medium-

sized and large cities (population over 20,000) were relat-

ed to the urban municipality’s population size. Figure 5

shows a clear logarithmic relationship of medium-sized

cities being more compact than the big ones.

A similar relation analysis was conducted for the urban

population density (Fig. 6). Conversely, the urban popu-

lation density of bigger cities is higher, showing a clear

logarithmic trend.

Discussion

This research contributes to the study of urban compact-

ness in Poland, being the rst to employ a shape measure

to assess the compactness of all urban municipalities in the

country. Previous studies have predominantly focused on

a b

Compactness of Polish urban areas – methodologies and analysis based on CLC dataset 135

urban sprawl, often limited to a smaller scope of cities (for

example, Śleszyński 2014; Lityński 2021). Leveraging

GIS and information technology tools enabled the man-

agement of extensive data sets across numerous munici-

palities, providing a comprehensive analysis of all urban

areas over three distinct years for which spatial data was

available in CLC. This chapter discusses the ndings con-

cerning the primary research questions.

Methods for analysing the compactness

of Polish cities

The methodology utilised in this study combines the

widely used measure of urban population density with an

urban shape measure based on the assumption that a cir-

cle represents the most compact geometric shape. It uses

CLC- an open-sourced pan-European land use database in

combination with population data (GUS). The integration

of these measures proved eective in analysing compact-

ness at a national scale. The shape measure, in particular,

oers a novel approach in the Polish context, allowing

for a more nuanced understanding of urban form beyond

simple population density metrics. There is a considerable

overlap between compactness index and urban population

density measures concerning the geographic concentration

of the most compact and densest urban areas.

In addition, there is a strong correlation between the two

measures. Pearson’s correlation between Ud and Ci was

carried out for the three data sets. The results for all urban

municipalities (urban and urban-rural) for the 2006 data

showed an r-value correlation of 0.45 with a statistically

signicant p-value. In 2012, the correlation value increased

to 0.48. In the most recent period, the coecient remained

at 0.48. This strong relationship is related to the fact that

both measures have the same urban area component (Ad

in the compactness index and Ua in urban density). It also

Fig. 5. Urban compactness trend for medium-sized

and big urban municipalities in 2018

(elaborated by M. Bukowski)

Il. 5. Trend zwartości zabudowy w średnich i dużych

gminach miejskich w 2018 r.

(oprac. M. Bukowski)

Fig. 6. Urban population density trend for medium-sized

and big urban municipalities in 2018

(elaborated by M. Bukowski)

Il. 6. Trend gęstości zaludnienia obszarów miejskich

w średnich i dużych gminach miejskich w 2018 r.

(oprac. M. Bukowski)

proves that the measures represent the same urban phe-

nomenon.

Both measures exhibit a signicant decrease over time.

These observations conrm that these measures describe

a similar concept. Moreover, the presented outcomes of

density relation with size are in line with research con-

ducted by Śleszyński (2014), which showed a strong cor-

relation between the density of a city’s central area and

city size. However, it has to be noted that this measure,

while sucient for general nationscale research, is not

suciently detailed for detailed studies. This is because

there are discrepancies between land classes identied by

CLC and the actual land uses. For example, according to

the study of Śleszyński, Gibas and Sudra (2020), there is

a mismatch, especially in the urbanised areas classica-

tion. While this method’s alignment with existing literature

supports its validity, future improvement is required. Fur-

ther research should try to eliminate these discrepancies by

reaching for dierent data.

Relationship between city size

and compactness

The analysis reveals that compactness is more preva-

lent in medium-sized urban municipalities, whereas urban

population density is higher in larger urban municipali-

ties, with Warsaw being a notable outlier. However, there

is a si gnicant geographic overlap between the most com-

pact and densest urban areas, particularly in the southern

edge and above the central belt of Poland (Fig. 7). The

Kujawsko-Pomorskie voivodeship exhibits the most com-

pact and densest municipalities, with neighbouring regions

like Mazowieckie, Pomorskie, and northern parts of Lodz-

kie sharing this concentration. Conversely, lowland areas

show a sparse distribution of compact and dense cities,

while mountainous regions in the south and the lakelands

136 Ewa Szymczyk, Mateusz Bukowski

have higher concentrations. Moreover, the voivodeship

with sparse populations, such as the northeastern corner of

Po land, exhibits highdensity urban areas. These ndings

suggest that historical context and topography play crucial

roles in shaping urban compactness and pave the way for

future research on the causes of compact urban patterns.

Changes in compactness over time

This study provides evidence that the compactness of

Polish urban areas is declining over time. Both compact-

ness and urban population density measures demonstrate

signicant decreases, suggesting an ongoing trend towards

urban dispersion. This trend poses environmental and eco-

nomic risks, as highlighted by Lityński (2021), Śleszyń ski

(2014) and Schiller and Kenworthy (2018), among others,

and challenges the sustainability of urban development.

The implications are particularly concerning in the con-

text of declining demographic rates and urban shrinkage in

Poland (Szymczyk, Bukowski 2023). Shrinking cities nd

it harder to maintain excess infrastructure in low-density

areas. Research on shrinkage sprawl in the Polish context

should follow to keep track of urbanisation patterns to ef-

fectively manage this phenomenon’s negative eects.

Research limitations

It has to be noted that this study did not account for

topographical features such as bodies of water or forests,

which constrains urban area development (making it more

compact), nor did it consider the relationship to the CBD

or historic city cores (Ottensmann 2021). Moreover, this

study shows that the most compact cities are part of a big-

ger urban area with administrative boundaries “cutting”

the shape out. Such cities, e.g., Ząbki and Podkowa Leśna

around Warsaw, can be seen around big metropolitan areas.

It would not be fair to see them as separate compact enti-

ties in the same way as Kościan, a city which is not bor-

dering any big urban area, nor is it shaped by natural fea-

tures. This leads to the conclusion that urban delimitation

in future studies may try to use the concept of functional

areas instead of administrative boundaries. As dened by

the OECD (Moreno-Monroy, Schiavina and Veneri 2021),

functional urban areas (FUAs) take into account the con-

nections between cities and their surroundings. This may

help to capture suburban sprawl more accurately than pre-

sented in this study.

Moreover, this study acknowledges the discrepancies

between land uses, which tend to be overly generalised

in CLC. Future research could benet from incorporating

these factors and improving the accuracy. Exploring mor-

phological aspects like street connectivity building density

(Burton 2002) and combining it with economic and func-

tional aspects of compactness warrant further investigation.

Moreover, population data coming from municipal data

allowed for a generalised urban density calculation. This

could be improved by using population data with more

pre-

cise geolocation.

In conclusion, this study provides a robust framework

for analysing urban compactness in the European context,

oering valuable insights into the spatial dynamics of ur-

ban areas, such as the spatial implications of planning pol-

icies or economic processes. Since CLC datasets cover all

European Union member states, this method is replicable

in every member state, allowing for further comparative

studies. The ndings underscore the need for tailored ur-

ban policies that address the challenges of urban dispersion

and promote sustainable urban development.

Summary

This study contributes to the state of knowledge by intro -

ducing and testing a shape measure to assess urban com pact-

ness across urban areas, oering a general but com prehen

sive national analysis. The research reveals that compactness

in Poland is falling over the analysed time between 2006 and

2018. It shows that smaller urban mu ni cipalities are more

compact in shape, whereas larger municipalities exhibit a lo-

garithmic pattern of higher urban population density. Geo-

graphic patterns indicate that compact cities are more con-

centrated in Poland’s southern and central regions.

Fig. 7. The compactness index (left)

and urban population density calculation (right)

for all Polish municipalities in three periods

based on CLC datasets and Polish Statistics

(elaborated by M. Bukowski, E. Szymczyk)

Il. 7. Indeks zwartości (po lewej) oraz obliczenia gęstości zaludnienia

obszarów miejskich (po prawej) dla wszystkich polskich gmin

w trzech okresach, na podstawie danych CLC i statystyk GUS

(oprac. M. Bukowski, E. Szymczyk)

Compactness of Polish urban areas – methodologies and analysis based on CLC dataset 137

Despite these advancements, the measure should have

considered factors such as functional urban areas, to-

pography, the location of a city in relation to other cities

(free standing or adjacent) and the aspects of proximity

of urban areas to city centres. Future research should also

incorporate morphological factors like street connectivity,

building density, building typologies, and economic and

functional aspects of compactness. Addressing these lim-

itations will enhance understanding of urban dynamics and

inform more eective urban planning and policymaking,

contributing to more sustainable development patterns in

the context of climate change as well as in times of demo-

graphic and urban shrinkage.

References

Ahlfeldt, Gabriel, and Elisabetta Pietrostefani. “The compact city in em-

pirical research: A quantitative literature review”. SERC Discussion

Paper, 215 (2017). Accessed March 10, 2024, at https://eprints.lse.

ac.uk/83638/1/sercdp0215.pdf.

Altman, Micah. “Districting Principles and Democratic Representation.”

Ph.D thesis. California Institute of Technology, 1998. https://doi.org/

10.7907/7ZE9-TH19.

Barnes, Richard, and Justin Solomon. “Gerrymandering and Compact-

ness: Implementation Flexibility and Abuse.” Political Analysis 29,

no. 4 (2021): 448–66. https://doi.org/10.1017/pan.2020.36.

Burton, Elisabeth. “Measuring Urban Compactness in UK Towns and

Cities.” Environment and Planning B: Planning and Design 29, no. 2

(2002): 219–50. https://doi.org/10.1068/b2713.

Chambers, Christopher P., and Allan D. Miller. “A Measure of Bizarre-

ness.” Quarterly Journal of Political Science 5, no. 1 (2010): 27–44.

http://dx.doi.org/10.1561/100.00009022.

CLC. “Corine Land Cover Data.” Accessed March 20, 2023. https://land.

copernicus.eu/en/products/corine-land-cover.

Dantzig, George B., and Thomas L. Saaty. Compact City: A Plan for

a Liveable Urban Environment. WH Freeman, 1973.

GUGiK. “Geodesy and Cartography Data.” Accessed March 20, 2024, at

https://www.gov.pl/web/gugik-en.

GUS BDL. 2023. “Statistics Poland Data.” Accessed March 1, 2023, at

https://bdl.stat.gov.pl/bdl/start.

Lityński, Piotr. “The Intensity of Urban Sprawl in Poland.” ISPRS Inter-

national Journal of Geo-Information 10, no. 2 (2021): 95. https://doi.

org/10.3390/ijgi10020095.

Moreno-Monroy, Ana, Marcello Schiavina, and Paolo Veneri. “Metropol-

itan Areas in the World: Delineation and Population Trends.” Journal

of Urban Economics 125 (2021): 103242. https://doi.org/10.1016/j.

jue.2020.103242.

Mubareka, Sarah, Eric Koomen, Christine Estreguil, and Carlo Lavalle.

“Development of a composite index of urban compactness for land

use modelling applications.” Landscape and Urban Planning 103,

no. 3–4 (2011): 303–17. https://doi.org/10.1016/j.landurbplan. 2011.

08.012.

Newman, Peter, and Jerey Kenworthy. Cities and Automobile Depend-

ence: An International Sourcebook. Gower, 1989.

Niemi, Richard G., Bernard Grofman, Carl Carlucci, and Thomas Hofeller.

“Measuring Compactness and the Role of a Compactness Standard

in a Test for Partisan and Racial Gerrymandering.” The Journal of

Politics 52, no. 4 (1990): 1155–81. https://doi.org/10.2307/2131686.

OECD. “Compact City Policies A Comparative Assessment.” Pub-

lished 2012. Accessed March 20, 2024, at https://www.oecd.org/

greengrowth/ compact-city-policies-9789264167865-en.htm.

Ottensmann, John. Measuring the Shape of Urban Areas. Published Feb-

ruary, 2021. Accessed March 20, 2024, at https://doi.org/10.13140/

RG.2.2.28509.13288.

Reis, José P., Elisabete A. Silva, and Paulo Pinho. “Spatial metrics to

study urban patterns in growing and shrinking cities.” Urban Geog-

raphy 37, no. 2 (2015): 246–71. https://doi.org/10.1080/02723638.2

015.1096118.

Schiller, Preston, and Jerey Kenworthy. An Introduction to Sustainable

Trans portation: Policy, Planning and Implementation. Earthscan, 2018.

https://doi.org/10.4324/9781315644486.

Schwarz, Nina, Dagmar Haase, and Ralf Seppelt. “Omnipresent Sprawl?

A Review of Urban Simulation Models with Respect to Urban

Shrinkage.” Environment and Planning B: Planning and Design 37,

no. 2 (2010): 265–83. https://doi.org/10.1068/b35087.

Schwartzberg, Joseph E. “Reapportionment, Gerrymanders, and the No-

tion of Compactness.” Minnesota Law Review 50 (1966): 443–52.

Siedentop, Stefan, and Stefan Fina. “Urban Sprawl beyond Growth:

the Eect of Demographic Change on Infrastructure Costs.” Flux,

no. 79–80 (2010): 90–100. https://doi.org/10.3917/ux.079.0090.

Song, Yu, Guofan Shao, Xiaodong Song, Yong Liu, Lei Pan, and Hong

Ye. “The Relationships between Urban Form and Urban Commuting:

An Empirical Study in China.” Sustainability 9, no. 7 (2017): 1150.

https://doi.org/10.3390/su9071150.

Squires, Gregory D., ed. Urban Sprawl: Causes, Consequences, and Pol-

icy Responses. The Urban Institute Press, 2002.

Szymczyk, Ewa, and Mateusz Bukowski. “Identication of Shrink-

ing Cities in Poland Using a Multi-Criterion Indicator.” Przegląd

Geo graczny 95, no. 4 (2023): 447–73. https://doi.org/10.7163/

PrzG. 2023.4.5.

Śleszyński, Przemysław. “Distribution of population density in Polish

towns and cities.” Geographia Polonica, no. 1 (2014): 61–75. http://

dx.doi.org/10.7163/GPol.2014.4.

Śleszyński, Przemysław, Piotr Gibas, and Paweł Sudra. “The Problem

of Mismatch between the CORINE Land Cover Data Classication

and the Development of Settlement in Poland.” Remote Sensing 12,

no. 14 (2020): 2253. https://doi.org/10.3390/rs12142253.

Thinh, Nguyen Xuan, Guenter Arlt, Bernd Heber, Joerg Hennersdorf, and

Iris Lehmann. “Evaluation of urban land-use structures with a view

to sustainable development.” Environmental Impact Assessment Re-

view 22, no. 5 (2002): 475–92. https://doi.org/10.1016/S0195-9255-

(02) 00023-9.

Tsai, YuHsin. “Quantifying Urban Form: Compactness Versus ‘Sprawl’.”

Urban Studies 42, no. 11 (2005): 141–61. https://doi.org/ 10. 1080/

00420980420003097.

“Uchwała nr 198 Rady Ministrów z dnia 20 października 2015 r. w spra-

wie przyjęcia Krajowej Polityki Miejskiej.” MP 2015 poz. 1235.

Accessed January 15, 2025, at https://isap.sejm.gov.pl/isap.nsf/Doc-

Details.xsp?id=wmp20150001235.

UN-HABITAT. “Envisaging the Future of Cities.” Published 2022. Ac-

cessed March 20, 2024, at https://unhabitat.org/world-cities-report-

2022-envisaging-the-future-of-cities.

138 Ewa Szymczyk, Mateusz Bukowski

Streszczenie

Kompaktowość polskich terenów zurbanizowanych

– metody badań i analiza z użyciem bazy danych CLC

W artykule autorzy przedstawili dynamikę kompaktowości (zwartości) zabudowy miejskiej w Polsce, analizując metody oceny tego zjawiska

oraz zmiany zachodzące w czasie. Trzy główne pytania badawcze skoncentrowali wokół metod analizy zwartości, związku pomiędzy wielkością

gmin miejskich a zwartością oraz zmian czasowych. Badaniem objęto wszystkie polskie gminy i zanalizowano wybrane miary zwartości w latach

2006, 2012 i 2018. Poprzez ocenę różnych metod analitycznych oraz wykorzystanie danych o użytkowaniu terenu z bazy CORINE Land Cover au-

torzy zaproponowali nową metodologię monitorowania wskaźników zwartości miast, wspierając tym samym podejmowanie decyzji planistycznych

opartych na danych, dążąc do zrównoważonego planowania przestrzennego.

Wyniki ujawniają wzorce i korelacje pomiędzy wielkością miast a zwartością w czasie, wskazując na ogólny spadek zwartości zabudowy w ba-

danym okresie (2006–2018). Wykazano istnienie logarytmicznego związku między wielkością populacji, indeksem zwartości i gęstością zaludnienia

obszarów miejskich. Mniejsze miasta cechują się większą zwartością kształtu, ale niższą gęstością, natomiast większe miasta są gęściej zaludnione,

lecz jednocześnie bardziej rozproszone. Praca wnosi wkład do stanu wiedzy, wprowadzając i testując metodę oceny zwartości zabudowy miejskiej

w różnych typach gmin. Ponieważ dane CLC obejmują cały obszar Unii Europejskiej, metoda ta może być replikowana w każdym państwie człon-

kowskim, umożliwiając dalsze badania porównawcze.

Słowa kluczowe: Polska, zwartość, kompaktowość, gęstość zaludnienia obszarów miejskich, CLC, metryki formy urbanistycznej