Satellite Monitoring of Urban Growth

and Indicator-based Assessment of

Environmental Impact
Dorothy Furberg

Licentiate Thesis in Geoinformatics

Royal Institute of Technology (KTH)
Department of Urban Planning and Environment
100 44 Stockholm

TRITA SoM 2014-15

ISSN 1653-6126
ISRN KTH/SoM/2014-15/SE
ISBN 978-91-7595-353-3

Dorothy Furberg

Printed by
Universitetsservice US AB
Stockholm, Sweden, 2014

ii

Abstract

One of the major consequences of urbanization is the transformation of land

surfaces from rural/natural environments to built-up land that supports
diverse forms of human activity. These transformations impact the local
geology, climate, hydrology, flora and fauna and human-life supporting
ecosystem services in the region. Mapping and analysis of land use/land
cover change in urban regions and tracking their environmental impact is
therefore of vital importance for evaluating policy options for future growth
and promoting sustainable urban development.
The overall objective of this research is to investigate the extent of urban
growth and/or sprawl and its potential environmental impact in the regions
surrounding a few selected major cities in North America, Europe and Asia
using landscape metrics and other environmental indicators to assess the
landscape changes. The urban regions examined are the Greater Toronto
Area (GTA) in Canada, Stockholm region and County in Sweden and
Shanghai in China. The analyses are based on classificatons of optical
satellite imagery (Landsat TM/ETM+ or SPOT 1/5) between 1985 and 2010.
Maximum likelihood classification (MLC) under urban/rural masks, objectbased image analysis (OBIA) with rule-based classification and support
vector machines (SVM) classification methods were used with grey level cooccurrence matrix (GLCM) texture features as input to help obtain higher
accuracies. Based on the classification results, landscape metrics, selected
environmental indicators and indices, and ecosystem service valuation were
calculated and used to estimate environmental impact of urban growth.
The results show that urban areas in the GTA grew by nearly 40% between
1985 and 2005. Results from the landscape metrics and urban compactness
indicators show that low-density built-up areas increased significantly in the
GTA between 1985 and 2005, mainly at the expense of agricultural areas.
The majority of environmentally significant areas were increasingly
surrounded by urban areas between 1985 and 2005, furthering their isolation
from other natural areas. Urban areas in the Stockholm region increased by
10% between 1986 and 2006. The landscape metrics indicated that natural
areas became more isolated or shrank whereas new small urban patches
came into being. The most noticeable changes in terms of environmental
impact and urban expansion were in the east and north of the study area.
Large forested areas in the northeast dropped the most in terms of
environmental impact ranking, while the most improved analysis units were
close to the central Stockholm area. The study comparing Shanghai and

iii

Stockholm County revealed that urban areas increased ten times as much in
Shanghai as they did in Stockholm, at 120% and 12% respectively. The
landscape metrics results show that fragmentation in both study regions
occurred mainly due to the growth of high density built-up areas in
previously more natural environments, while the expansion of low density
built-up areas was for the most part in conjunction with pre-existing patches.
The growth in urban areas resulted in ecosystem service value losses of
approximately 445 million USD in Shanghai, mostly due to the decrease in
natural coastal wetlands, while in Stockholm the value of ecosystem services
changed very little.
This study demonstrates the utility of urban and environmental indicators
derived from remote sensing data via GIS techniques in assessing both the
spatio-temporal dynamics of urban growth and its environmental impact in
different metropolitan regions. High accuracy classifications of optical
medium resolution remote sensing data are achieved thanks in part to the
incorporation of texture features for both object- and pixel-based
classification methods, and to the use of urban/rural masks with the latter.
The landscape metrics calculated based on the classifications are useful in
quantifying urban growth trends and potential environmental impact as well
as facilitating their comparison. The environmental indicator results
highlight the challenges in terms of sustainable urban growth unique to each
landscape, both spatially and temporally. The next phase of this PhD
research will involve finding valid methods of comparing and contrasting
urban growth patterns and estimated environmental impact in different
regions of the world and further exploration of how to link urbanizing
landscapes to changes in ecosystem services via environmental indicators.

Keywords: Urban growth, remote sensing, landcover classification,

landscape metrics, environmental indicators, environmental impact, Greater
Toronto Area, Stockholm, Shanghai

iv

Acknowledgements
This study was supported by the Swedish Research Council for
Environment, Agricultural Sciences and Spatial Planning (FORMAS). This
research is also part of the project Satellite Monitoring of Urbanization for
Sustainable Urban Development within the European Space Agency (ESA)
and Chinese Ministry of Science and Technology (MOST) Dragon 2
program.
I would first of all like to thank my supervisor, Professor Yifang Ban, for
giving me the opportunity to pursue doctoral studies in Geoinformatics. I am
very grateful for her guidance, instruction, resourcefulness, support and
flexibility over the years, especially when circumstances change.
I would also like to thank Jan Haas for a great teamwork experience as we
studied/compared Stockholm and Shanghai and for his hard work on the
article. Thank you to past and present staff at Geoinformatics for good
discussions, exchange and comradery. Thanks especially to Alexander Jacob
for current (and sometimes long distance) tech-support and assistance and to
Irene Rangel for past tech-support and cheerful instruction. Many thanks
also go to Hans Hauska for his guidance, support and helpfulness throughout
the years.
In addition, I would like to thank an anonymous reviewer of the GTA paper.
Their thoughtful questions and constructive critique were very instructive
and helped to improve the quality of that paper. They were regrettably not
mentioned in the acknowledgments of the article and so I take the
opportunity here.
Finally, I am very grateful for my family and especially to my husband,
Richard, who has given much to support my studies and has been flexible,
patient and a voice of reason. I could not have come this far without him.
Dorothy Furberg
Stockholm, November 2014

Table of contents
Abstract
Acknowledgements

iii
v

1 Introduction ................................................................................ 13
1.1 Rationale ............................................................................... 13
1.2 Research objectives ............................................................... 14
1.3 Thesis structure ..................................................................... 15
1.4 Statement of contributions ..................................................... 16
2 Background and literature review ............................................ 17
2.1 Classification techniques of medium- and high-resolution
multispectral remote sensing data over urban areas ..................... 17
2.1.1 Pixel-based classification ............................................... 19
2.1.2 Object-based image analysis and classification ............. 20
2.2 Combining indicators and remote sensing/GIS for assessment
of environmental impact and landscape change monitoring ......... 22
2.2.1 Landscape metrics as indicators of landscape change and
environmental impact ................................................................. 24
2.2.2 Remote sensing and the use of ecosystem service
indicators .................................................................................... 25
3 Study areas and data description .............................................. 27
3.1 Greater Toronto Area............................................................ 27
3.2 Stockholm region and County ............................................... 28
3.3 Shanghai ................................................................................ 31
4 Methodology ............................................................................... 32
4.1 Image processing................................................................... 32
4.1.1 Image pre-processing ..................................................... 32
4.1.2 Texture analysis with Grey-Level-Co-occurrence-Matrix
features ....................................................................................... 32
4.2 Image classification............................................................... 33
4.2.1 Maximum Likelihood Classification .............................. 33

vi

4.2.2 Object-based Image Analysis and rule-based

classification ............................................................................... 34
4.2.3 Support Vector Machines classification ......................... 34
4.3 Accuracy assessment ............................................................. 34
4.4 Landscape metrics ................................................................. 35
4.5 Urban and environmental indicators and indices ................. 38
4.5.1 GTA ................................................................................ 38
4.5.2 Stockholm region ........................................................... 39
4.5.3 Stockholm and Shanghai comparison ............................ 41
4.6 Valuation of Ecosystem Services ........................................... 42
5 Results and discussion ................................................................ 42
5.1 Image classifications ............................................................. 42
5.2 Landscape change and potential environmental impact
analysis using landscape metrics ................................................... 47
5.2.1 Urban growth and landscape change trends ................... 47
5.2.2 GTA ................................................................................ 49
5.2.3 Stockholm city and region.............................................. 50
5.2.4 Stockholm County and Shanghai ................................... 51
5.3 Assessing impact with urban form, environmental and
ecosystem service indicators .......................................................... 52
5.3.1 GTA ................................................................................ 52
5.3.2 Stockholm region ........................................................... 53
5.3.3 Stockholm County and Shanghai comparison................ 54
5.4 Comparison of the GTA, Stockholm and Shanghai
investigations .................................................................................. 55
6 Conclusions and future research .............................................. 58
6.1 Conclusions ........................................................................... 58
6.2 Future research ..................................................................... 60
References ......................................................................................... 63
Appendix ........................................................................................... 82

vii

List of figures
Figure 1 The DPSIR framework for Reporting on Environmental
Issues .......................................................................................... 23
Figure 2 2005 Landsat TM images and study extent of the Greater
Toronto Area .............................................................................. 28
Figure 3 Study areas from Papers I, III and IV ................................ 30
Figure 4 Flowchart comparing image processing and classification
methodologies for each study area ............................................. 33
Figure 5 Flowchart comparing environmental indicator calculations
for each study area ...................................................................... 35
Figure 6 Land cover classification results for the GTA in 1985 and
2005 ............................................................................................ 44
Figure 7 Land cover classification results for Stockholm region in
1986 and 2006 ............................................................................ 45
Figure 8 Land cover classification results for Shanghai and
Stockholm County in 1989/90, 2000 and 2010 .......................... 46
Figure 9 Growth trends for urban vs. non-urban land cover in the
GTA, Shanghai, Stockholm County and region during the period
1985-2010................................................................................... 48
Figure 10 Contagion trends for the GTA, Stockholm County and
Shanghai during the period 1985-2010 ...................................... 49

viii

List of tables
Table 1 Compiled set of landscape metrics used in the current
research....................................................................................... 35
Table 2 Table indicating in which studies each LM was used ........ 37
Table 3 Urban indicators used in the GTA study ............................ 39
Table 4 Enviornmental impact indicator specifications used in Paper
III ................................................................................................ 40
Table 5 Comparison of overall classification accuracies and kappa
coefficients for all classifications performed ............................. 43

ix

List of abbreviations
AHP
ANN
AVHRR
AWDI
CART
CICES

- Analytic Hierarchy Process

- Artificial Neural Network
- Advanced Very High Resolution Radiometer
- Average-Weighted Distance Indicator
- Classification and Regression Tree
- Common International Classification of Ecosystem
Services
CRDA
- Co-central Ring Density Analysis
CWED
- Contrast-Weighted Edge Density
DEM
- Digital Elevation Model
DPSIR
- Driving forces, Pressure, State, Impact, Response
EI(A)
- Environmental Impact (Assessment)
ES
- Ecosystem Services
ESA
- Environmentally Significant Area
ESV
- Ecosystem Service Value
(E)TM
- (Enhanced) Thematic Mapper
GLCM
- Gray-Level Co-occurrence Matrix
GLS
- Global Land Survey
GTA
- Greater Toronto Area
HDB
- High-Density Built-up
ISODATA - Self-Organizing Data Analysis Technique Algorithm
KNN
- K-Nearest Neighbors
LCA
- Landscape Change Analysis
LDB
- Low-Density Residential Built-up
LFA
- Large Forested Area
LM
- Landscape Metrics
LULC
- Land Use/Land Cover
MLC
- Maximum Likelihood Classifier
MODIS
- Moderate Resolution Imaging Spectroradiometer
OBIA
- Object-Based Image Analysis
ORM
- Oak Ridges Moraine
RF
- Random Forests

SPOT
SVM
TRCA
USGS
VHR

- Satellite Pour lObservation de la Terre

- Support Vector Machines
- Toronto Region and Conservation Authority
- United States Geological Survey
- Very High Resolution

xi

1 Introduction
1.1 Rationale
Urbanization poses numerous challenges for those working towards
sustainable development. While cities may experience internal problems as
they grow, their impact on the surrounding external natural environment
upon which they depend is of critical importance. Lambin et al. (2001)
describe this relationship as follows:
In reality, urbanization affects land change elsewhere through the
transformation of urban-rural linkages. For example, urban inhabitants
within the Baltic Sea drainage depend on forest, agriculture, wetland,
lake and marine systems that constitute an area about 1000 times
larger than that of the urban area proper (Folke et al., 1997). Given
that urban life-styles tend to raise consumption expectations and that
60% of the worlds population will be urban by 2025 (United Nations
Population Fund, 1991), the ruralurban linkage or the urban
ecological footprint is critical to land change assessments.
Mapping and analysis of land use/land cover (LULC) change in urban
regions is therefore crucial to tracking this ecological footprint and
deciding policy options and/or remedies for future growth and environmental
conservation. Wentz et al. (2009) also emphasize the importance of this task:
Urbanization represents one of the most significant alterations that
humankind has made to the surface of the Earth It is essential that we
document, to the best of our ability, the nature of land transformations and
the consequences to the existing environment.
Past research has demonstrated the utility of remote sensing data and
geographic information systems (GIS) to capture and map urban LULC
change (e.g., Cihlar 2000; Barnsley and Barr 2000; Franklin and Wulder
2002; Zhou et al. 2008; Yang 2011; Qin et al. 2013; Ban et al. 2014a;
2014b). Tracking the environmental impact of these changes has often been
undertaken with the help of landscape metrics (Forman and Godron 1986;
ONeill et al. 1988; Turner 1990; Haines-Young et al. 1993; Hargis et al.
1998; Botequilha Leitao and Ahern 2002; and e.g., McGarigal and McComb
1995; Narumalani et al. 2004; Kamusoko and Aniya 2007; Li et al. 2010;
Haas and Ban 2014). But the use of environmental indicators in conjunction
with remote sensing and GIS techniques for estimation of potential

13

environmental impact in urban regions is an underexplored area of research.

There is a need for development of ecosystem condition indicators from GIS
and remote sensing data for mitigation and planning purposes (Revenga
2005). Aplin (2005) points out that remote sensing specialists have perhaps
focused on technological issues as their principal concern, rather than
ecological problems. Newton et al. (2009) note that there exists a greater
potential for use of remote sensing within landscape ecology but also draw
attention to a traditional divide between the remote sensing and ecological
science research communities. This research attempts to address important
aspects in both fields by exploring the utility of selected environmental
indicators for impact assessment derived from classified remote sensing data
in selected urban regions in Europe, North America and Asia. By comparing
patterns and impact of urban growth in different regions, the results would
ideally provide valuable information, ideas and tools for those involved in
urban and environmental planning in many different locations.

1.2 Research objectives

The overall objective of this research is to investigate the extent of urban
growth and/or sprawl and its potential environmental impact in the regions
surrounding a few selected major cities in North America, Europe and Asia
using landscape metrics and other environmental indicators to assess the
landscape changes.
The primary scientific question is:
(1) When and where has urban growth/sprawl occurred in these cities over
the past few decades and what are the probable impacts on the
surrounding agricultural/natural environment as a result of the urban
expansion?
A secondary scientific question is:
(2) Is there a widely applicable approach to evaluating probable
environmental impact from urbanization in metropolitan regions based
on optical remote sensing data and the use of indicators, and is it
possible to compare patterns of urban growth and environmental
impact between cities?

14

The specific research objectives are:

(1) To monitor urban growth over the past few decades in Toronto,
Canada, Stockholm, Sweden and Shanghai, China using optical
remote sensing imagery.
(2) To analyze landscape change in and around the cities and to evaluate
probable environmental impact from urban growth using information
from landscape metrics and other tailored environmental indicators.
(3) To evaluate the methods used in terms of finding a widely applicable
approach to estimating environmental impact from urbanization in
metropolitan landscape regions based on optical remote sensing data.

1.3 Thesis structure

The thesis is structured as follows: Chapter 1 gives an overview of the
research, including the background, research objectives and organisation of
the thesis. Chapter 2 reviews the state-of-the-art of research in terms of
classification of medium to high resolution optical satellite imagery and the
use of landscape metrics and indicators for assessing environmental impact
in urban areas. Chapter 3 describes the study areas and data used for the
research and Chapter 4 outlines the methodologies employed. Chapter 5
presents and compares the results from the study areas. Chapter 6 draws
conclusions and discusses the potential for future research.
This thesis is based on the following papers, which are referred to in the text
by their Roman numerals.
I. Furberg, D. and Ban, Y., 2010. Satellite monitoring and impact assessment
of urban growth in Stockholm, Sweden between 1986 and 2006. In Imagin
[e, g] Europe: Proceedings of the 29th Symposium of the European
Association of Remote Sensing Laboratories, Chania, Greece: 131-142. IOS
Press.
II. Furberg, D. and Ban, Y., 2012. Satellite Monitoring of Urban Sprawl and
Assessment of its Potential Environmental Impact in the Greater Toronto
Area between 1985 and 2005. Environmental management 50(6): 10681088.
III. Furberg, D. and Ban, Y., 2013. Satellite Monitoring of Urban Land
Cover Change in Stockholm Between 1986 and 2006 and Indicator-Based

15

Environmental Assessment. In Earth Observation of Global Changes

(EOGC): 205-222. Springer Berlin Heidelberg.
IV. Haas, J., Furberg, D. and Ban, Y., 2014. Satellite Monitoring of
Urbanization and Environmental Impacts: A Comparison of Stockholm and
Shanghai. Submitted to International Journal of Applied Earth Observation
and Geoinformation (revised and re-submitted).

1.4 Statement of contributions

Paper I
All analyses and methodologies of paper I were developed and performed by
the main author under the supervision of Professor Ban, the 2nd author.
Professor Ban initiated the ideas for this paper and has been involved in the
development of the paper.
Paper II
All analyses and methodologies of paper II were developed and performed
by the main author under the supervision of Professor Ban, the 2nd author.
Professor Ban initiated the ideas for this paper and has been involved in the
development of the paper.
Paper III
All methodologies and analyses of paper III were developed and performed
by the main author under the supervision of Professor Ban, the 2nd author,
who initiated the ideas for these papers and has been involved in their
development.
Paper IV
Professor Ban, the 3rd author, proposed the topic for this paper. Methodology
development was performed by the first author together with the second
author under the supervision of professor Ban. Study area description, image
processing, classifications, post-processing, accuracy assessment, landscape
metric analysis and the discussion part for Shanghai were performed by the
first author and for Stockholm by the second author, with the exception of
the SVM classification which was performed by a departmental colleague,
Martin Sjstrm. Urbanization indices and ES were calculated by the first
author. The abstract, introduction and data description parts were mainly
written by the first author with editorial input from the second author. The
selection and interpretation of LM are mainly based on the knowledge and
previous research experience of the second author.

16

2 Background and literature review

In recent decades, numerous studies have made use of remote sensing data
over urban areas for landscape change analysis. Herold et al. (2005) have
pointed out that the combination of remote sensing and spatial metrics data
provide more detailed and spatially consistent information on urban structure
and change than either technique can separately. Revenga (2005) has
highlighted the need for development of ecosystem condition indicators from
GIS and remote sensing data for mitigation and planning purposes. The
combined use of these tools has great potential to further our knowledge of
landscape/use change and the results derived could influence and improve
urban planning. The following sections describe the research context and
recent developments in terms of classification of optical remote sensing data
over urban areas and the use of indicators, including landscape metrics, to
monitor landscapes and assess environmental impact. The derivation of
meaningful metrics and indicators is dependent upon the generation of wellclassified land cover/land use maps from remote sensing data, therefore
classification techniques are the focus of the first section.

2.1 Classification techniques of medium- and

high-resolution multispectral remote sensing
data over urban areas
Remote sensing data is a valuable source of information for the study of
urban areas. It can be obtained over large regions with good spatial
consistency and geometric detail at high temporal frequency (Herold et al.
2005). Thematic information such as land cover/land use change can be
obtained from remote sensing data once it has been classified. LULC
classification of multispectral remote sensing data based on statistical pattern
recognition techniques is one of the most commonly used methods of
information extraction (Narumalani et al. 2002).
A number of classification methods are available in order to generate landcover/use maps from optical remote sensing data, including algorithms based
on parametric and nonparametric statistics, nonmetric methods, supervised
or unsupervised classification logic, hard or fuzzy set classification logic,
per-pixel or object-oriented classification logic and hybrid approaches
(Jensen 2005). None of them are absolutely superior to the others; all
classifiers are subject to a three-way compromise between the spectral
information content of the imagery, the method of making class decisions

17

and the information classes that are desired (Franklin and Wulder 2002). The
choice of classification method will depend on physical characteristics and
prior knowledge of the study area, the distribution of the remote sensing data
and the nature of the classification problem itself.
The focus of this research is on a regional metropolitan scale (thousands of
square kilometers) in order to take into account the whole of an urban area
and its surrounding natural environment. This is also an important extent for
assessing environmental impact since detrimental effects of fragmentation
on biodiversity and ecosystem services are generally found at intermediate
(regional) spatial scales (Olff and Ritchie 2002). While there have been
numerous studies conducted on a local, city or district level (Alberti and
Marzluff 2004; Lvenhaft et al. 2004; Mrtberg et al. 2007; Bino et al. 2008;
Li et al. 2010) and the global, continental or national level (Petit et al. 2001;
Gerard et al. 2005; Halada et al. 2009; Sthl et al 2011; Seto et al. 2012;
Gneralp and Seto 2013), there is a lack of studies on a regional
metropolitan level linking urbanization to its region-specific environmental
impact, and more that include a landscape perspective are needed
(Lfvenhaft et al. 2004; Seto et al. 2012). This research examines study sites
at this particular scale.
The choice of remote sensing data must be well-suited to the spatial extent of
the study area (Ban et al. 2014b). Low-resolution, large extent data such as
MODIS or AVHRR work well for studies over continental or global regions.
VHR data such as Quickbird or Ikonos favor more detailed, smaller-scale
studies, such as at the sub-city level or for examining specific habitat types.
For an intermediate regional approach, medium- to high-resolution satellite
data offers the best compromise between spatial coverage and level of detail
and is therefore employed.
The three classification methods used in this research - supervised pixelbased classification using the maximum likelihood classifier (MLC)
algorithm with texture as well as spectral data inputs (parametric), objectand knowledge-based classification (non-parametric), and pixel-based
support vector machines (SVM) classification (non-parametric) also with
textural inputs - were selected based on appropriateness for the study area,
degree of prior knowledge and remote sensing data available, as well as the
experience and degree of success of other authors in using them to conduct
studies of LULC change over urban areas.

18

2.1.1 Pixel-based classification

Pixel-based unsupervised classification techniques are arguably best suited
to large-area land cover classifications if the study area is not well-known,
due to the tremendous amount of training data otherwise required (Cihlar
2000). Supervised classification presents a better alternative when there is a
defined area to be studied and when medium resolution satellite data and
prior knowledge of the site are available. Landsat data has historically been
important for many different types of land cover assessments including
monitoring of urban areas, but its usage recently increased significantly since
the USGS Landsat archive was made freely available in 2009 (Weng et al.
2014). Herold (2009) notes the usefulness of Landsat data for detection of
land cover configuration within urban environments.
Many supervised algorithms have been developed to perform pixel-based
classification. Li et al. (2014) recently tested and compared 13 of these,
including MLC, KNN, CART, RF, SVM and machine learning algorithms,
by classifying Landsat data over an urban area in China. They found that
most of the algorithms performed well given sufficiently respresentative
training samples. MLC proved to be the most robust algorithm in that it
required the least amount of training data in order to achieve one of the
highest accuracies. Several studies have demonstrated the basic utility of the
MLC classifier when used on Landsat TM imagery (e.g., Wakelyn 1990; Lo
1998; Weng 2002; Kamusoko and Aniya 2007). However, more studies have
shown that the results obtained from MLC can be significantly improved
when combined with other techniques and data inputs (e.g., Hansen et al.
2001; Liu et al. 2002). Rozenstein and Karnieli (2011) found that using a
hybrid classification approach with ISODATA (unsupervised algorithm) and
MLC yielded better results than using the supervised classifier alone. Herold
et al. (2007) have noted that additional information, such as textural, spatial
or contextual, is often required to aid in successfully discriminating spectral
signals for mapping urban land-use types. Lu and Weng (2005) found that
the addition of higher resolution fused data and texture images improved
classification of Landsat ETM+ data over urban areas. Incorporation of
texture features in general have been found to improve image classification
accuracy, especially through the use of gray-level co-occurrence matrices
(GLCM) to derive them (Haralick et al. 1973), for Landsat and SPOT data
alike (Franklin and Peddle 1990; Gong et al. 1992; Butusov 2003;
Rodriguez-Galiano et al. 2012). This also holds true for classifications over
urban areas (Shaban and Dikshit 2001; De Martino et al. 2003; Herold et al.
2003).

19

A more recent group of supervised learning algorithms that perform

classification well with Landsat data are support vector machines (SVM),
originally introduced by Vapnik (1995) and Cortes and Vapnik (1995).
Unlike MLC, SVM are non-parametric classifiers which construct a
hyperplane in high-dimensional space with the largest possible distance to
the nearest training data point of any class. Since mapping in highdimensional space can be computationally heavy, a kernel function is
defined to suit the problem at hand. Mountrakis et al. (2011) reviewed
remote sensing applications of SVM and highlighted the advantages of their
ability to generalize well with limited training data and the lack of
requirement on underlying data distribution. SVM have been employed to
map land cover (e.g., Huang et al. 2002; Dixon and Candade 2008; Mathur
and Foody 2008;) and have been particularly successful in mapping urban
land cover (e.g., Huang et al. 2009; Hu and Ban 2012; Niu and Ban 2013).
Srivastava et al. (2012) tested various kernel functions and found in all cases
that SVM yielded consistently higher accuracy classifications of Landsat
imagery for LULC change investigation than did MLC. Dixon and Candade
(2008) compared classification of Landsat TM data using MLC, SVM and
Artificial Neural Network (ANN) and found that both ANN and SVM
outperformed MLC. SVM and ANN showed similar results in terms of
accuracy but the training time required by SVM was much less than for
ANN. Jin et al. (2005) tested SVM and MLC on textural features including
GLCM and found that SVM provided higher classification accuracy and
better generalization than MLC no matter which texture features were used.
Yet Li et al. (2014) found no advantage to using SVM over MLC in either a
pixel-based or object-based classification of Landsat data over an urban area.
They concluded that the quality and quantity of the training samples play a
bigger role in achieving high accuracy than the algorithm itself.

2.1.2 Object-based image analysis and classification

A number of studies have demonstrated the advantages of using object-based
image analysis (OBIA) over traditional pixel-based classification in urban
environments with medium- to high-resolution satellite imagery (Stefanov et
al. 2001; Wang et al. 2004; Cleve et al. 2008; Jacquin et al. 2008; Blaschke
2010; Ban and Jacob 2013; Jebur et al. 2013). Sometimes referred to as
image segmentation, OBIA can be defined as the division of an image into
spatially continuous, disjoint and homogeneous regions, also known as
objects based on color, shape and scale parameters. Apart from its main
advantage of using spatial information to create objects that more likely
correspond to actual structures or on-site areas than individual pixels do
(Ban et al. 2010), its decision rules tend to be dominated by the knowledge

20

of the human analyst rather than computer algorithms (Franklin and Wulder
2002) and there is the possibility of incorporating diverse types of data to
improve accuracy (Stefanov et al. 2001). Li et al. (2014) in their comparison
of 13 supervised algorithms found that object-based classification slightly
outperformed the pixel-based approaches in all instances based only on
spectral data. Wentz et al. (2009) point out the often higher classification
accuracy achieved with OBIA for urban areas thanks to its consideration of
shape, neighbourhood relations and contextual information. Darwish et al.
(2003) did a comparative study in which they tested object-based
classification against statistical classifiers using Landsat and IRS data over
urban areas. They showed that the object-oriented technique yields better
results with an increased number of land cover classes (11 vs. 5). On the
other hand, Duro et al. (2012) have found when comparing pixel- and objectbased approaches that they worked equally well when used with supervised
machine learning algorithms on SPOT data in agricultural environments.
More specifically, a number of studies have demonstrated the successful
application of OBIA to SPOT data for various purposes (e.g., Radoux and
Defourny 2007; Tiede et al. 2007; Conchedda et al. 2008; Su et al. 2008; Su
et al. 2009; Boggs 2010; Dimitrakopoulos et al. 2010; Lisita et al. 2013).
Tehrany et al. 2014 found that both of the object-based classification
approaches they tested (KNN and SVM) performed better than the pixelbased method (DT) for LULC mapping. Chen et al. (2009) made clear the
advantages of employing an object-oriented knowledge-based classification
method with SPOT 5 imagery in an urban environment over a pixel-based
approach. The former yielded the highest accuracy of several compared
methods and could create and distinguish between meaningful objects such
as roads and buildings, while providing a convenient way of incorporating
ancillary data such as DEM and textural information for the classification.
Jacquin et al. (2008) revealed OBIAs improved capacity to delineate urban
extent at regional scales with SPOT data. Newman et al. (2011)
demonstrated the advantages of an object-based approach over pixel-based
with regard to forest fragmentation assessment. Statistics Sweden (2008) has
used rule-based OBIA with SPOT 5 data for classification of urban green
areas and the Swedish Environmental Agency employs a rule-based objectoriented approach with SPOT data to classify nature types (Metria
GeoAnalys 2009).

21

2.2 Combining indicators and remote sensing/GIS

for assessment of environmental impact and
landscape change monitoring
Indicators are very useful as measurable criteria that point to the current
conditions of more complex phenomena. Aspinall and Pearson (2000) define
them as simple measures that represent key components of the system and
have meaning beyond the attributes that are directly measured.
Environmental indicators supply information on environmental problems
and can help identify key factors that cause pressure on the environment
(Smeets and Weterings 1999). They can thus help raise awareness of the
condition of ecosystems and assist policy makers in their planning decisions
(Revenga 2005). The DPSIR framework for reporting on environmental
issues (Driving forces, Pressure, State, Impact, Response) (Smeets and
Weterings 1999) shown in Figure 1 can be helpful in understanding human
influences on the surrounding natural environment and vice versa and in
classifying types of indicators as measures of different stages of this
interaction.
This framework shows the relationships between Driving forces and the
resulting environmental Pressures on the State of the Environment and
Impacts resulting from changes in environmental quality and on the societal
Response to these changes in the environment (Smeets and Weterings
1999). As will be seen, the indicators used in this research primarily measure
pressures on and the state of the environment in and around the study area
sites. There are many kinds of indicators that have been calculated to
measure pressure and state conditions of the environment. Examples within
the categories of biological, physical and chemical environmental indicators
include measures of air quality such as pollutant emissions and water
pollution such as eutrophication, changes in water temperature or incidence
of fish diseases. Socio-economic environmental indicators could include
measures of human-generated waste and human health measures such as
mortality and morbidity in relation to environmental quality (UNEP/RIVM
1994). Indicators must be carefully selected to fit the purpose of an
investigation or program and often a set of indicators is required to assess the
state of the environment (Niemeijer and de Groot 2008; van Oudenhoven et
al 2012).

22

Figure 1 The DPSIR framework for Reporting on Environmental Issues

(adapted from Smeets and Weterings 1999)
While many environmental indicators and indices have been developed over
the past several decades, their derivation from remote sensing and GIS data
is a newer field of research. Klemas (2001) notes that remote sensors can
monitor landscape level environmental indicators and that these become
particularly important as we shift to larger temporal, spatial and
organizational scales in order to study and compare the cumulative effects
of ecosystem degradation over entire landscapes and regions (Klemas
2001; Haines-Young et al. 1993). The body of research on environmental
indicators and indices derived from and/or used with remote sensing data
and GIS techniques is growing. Klemas (2001) showed how remote sensing
data can be used to detect landscape-level coastal environmental indicators,
specifically changes in LULC, riparian buffers and wetland condition.
Revenga (2005) highlighted useful examples of indicators for gauging
ecosystem conditions derived from GIS and remote sensing data. Nichol and
Wong (2007) assessed urban environmental quality using six different

23

parameters or indicators derived from a variety of satellite data; the

parameters were air temperature and quality, vegetation density, building
density and height, and noise. Krishnaswamy et al (2009) developed a multidate NDVI distance measure as a surrogate for forest type to measure its
variability on a single, continuous quantitative scale. Imhoff et al. (2010)
used impervious surface area from a Landsat-based dataset and land surface
temperature from MODIS to measure the urban heat island effect across a
range of biomes in the United States. Liang and Weng (2011) extracted
physical environmental variables from Landsat data and combined these
with socio-economic data to construct urban environmental quality indices
for a decadal comparison. Lakes and Kim (2012) evaluated the use of an
aggregated urban environmental indicator known as Biotope Area Ratio
together with classified remote sensing data for assessing and managing
urban ecosystem services. Michaud et al. (2014) used remotely sensed
environmental indicators such as the dynamic habitat index and snow cover
to extrapolate moose habitat in southern and central Ontario. De Sherbinin et
al. (2014) developed indicators derived from satellite data in three
categories: ambient air pollution, coastal eutrophication and biomass
burning. Behling et al. (2015) recently developed an automated GIS system
that can derive urban ecological indicators from hyperspectral remote
sensing data and height information.

2.2.1 Landscape metrics as indicators of landscape

change and environmental impact
An important concept established in the field of landscape ecology is that a
landscapes pattern strongly influences its ecological processes and
characteristics (Forman and Godron 1986; Turner 1989; McGarigal and
Marks 1995). Landscape fragmentation often negatively affects native
ecosystem function, and habitat fragmentation has been identified as one of
the greatest threats to biodiversity worldwide (Botequilha Leito and Ahern
2002; Leito et al. 2006; Lindenmayer and Fischer 2006) Landscape
fragmentation is often caused by conversion to human land uses such as
urban centers and transportation networks. Landscape metrics are wellestablished tools to measure change in landscape pattern and landscape
fragmentation in particular (ONeill et al. 1988; Turner 1990; Haines-Young
et al. 1993; Hargis et al. 1998; Botequilha Leito and Ahern 2002;
McGarigal 2002; Uuemaa et al. 2009). It is worth noting that landscape
metrics are also known as spatial metrics when taken out of the landscape
ecology context according to Herold et al. (2005) who define them as
measurements derived from the digital analysis of thematic-categorical
maps exhibiting spatial heterogeneity at a specific scale and resolution.

24

When applied to multi-temporal datasets, they can be used to describe and

analyze change in degree of spatial heterogeneity over time (Dunn et al.
1991) and constitute a very specific type of indicator useful for the
monitoring of landscape change.
A number of different sets of metrics, based on the work of ONeill et al.
(1988), have been developed, tested and revised (McGarigal and Marks
1995; Riitters et al. 1995; Hargis et al. 1998). However, many researchers
have urged caution in the use of landscape metrics, because they are often
strongly correlated and can be confounded (Li and Reynolds 1994;
McGarigal and Marks 1995; Botequilha Leitao and Ahern 2002; Leitao et al.
2006). Li and Wu (2004) point out the variable responses of certain
landscape indices to changes in classification scheme as well as the difficulty
in interpreting them since they often represent more than one aspect of
spatial pattern, the latter point having also been made by Li and Reynolds
(1994). They stress that simple metrics such as patch size, edge, inter-patch
distance and proportion are more likely to generate meaningful inferences.
Landscape or spatial metrics derived from remote sensing data have been
used to assess the impact of land-use changes on the environment (Uuemaa
et al. 2013; and e.g., Narumalani et al. 2004; Li et al. 2005; Kamusoko and
Aniya 2007; Long et al. 2010) and to characterize patterns of urban growth
(Reis et al. 2014; and e.g., Herold et al. 2003; Herold et al. 2005; Wu et al.
2011). There is a research potential to explore the dynamics between
changes in the environment in response to urban expansion with the help of
landscape metrics. One project that has undertaken this task is the Urban
Enviornmental Monitoring or 100 Cities Project (Wentz et al. 2009).
Research efforts in this specific area include Gao et al. (2012) who used
landscape indices to assess change in ecological security in nine cities in the
Pearl River Delta and Su et al. (2012 and 2014) who employed LM to guage
the impacts of urbanization on an eco-regional scale.

2.2.2 Remote sensing and the use of ecosystem

service indicators
Ecosystem services (ES) have been defined as the conditions and processes
through which natural ecosystems, and the species that make them up,
sustain and fulfill human life (Daily 1997). Studies examining their
definition, categorization and measurement have been underway since the
early 1990s (De Groot 1992; Costanza et al. 1997; Daily 1997). Various
attempts to map, model and assign measurable values to ecosystem services
at the landscape level have been the subject of more recent research (De

25

Groot et al. 2002; Troy and Wilson 2006; Willemen et al. 2008; Burkhard et
al. 2009; Nelson et al. 2009; Tallis and Polasky 2009). Development of a
standardized ES classification system has become a priority: the Millennium
Ecosystem Assessment (MA 2005) proposed a widely used classification
system of ES under the headings of supporting, regulating, provisioning and
cultural services and consultations organized by the European Environment
Agency are underway for the development of a common international
classification of ecosystem services or CICES (Haines-Young and Potschin
2013). Urban areas are heavily reliant on ES to meet the needs of the
populations they contain and yet the expansion of urban regions can have
serious detrimental effects on ecosystem services provided by the
surrounding natural landscape. Bolund and Hunhammar (1999) applied the
concept specifically to urban areas and Haase et al. (2014) have performed a
quantitative review of urban ES assessments dating up to 2012.
Remote sensing data has increasingly been used to gauge the extent and
condition of ecosystem services. Alcaraz-Segura et al. (2013) have reviewed
the use of remote sensing to quantify and monitor ES, particularly in relation
to carbon and water cycles, biodiversity and energy balance. Tallis et al.
(2012) have proposed a conceptual framework for global monitoring of
changes in ES based on remote sensing data in addition to other types of
regularly updated information. Feng et al. (2010) outlined the use of remote
sensing to monitor land cover, biodiversity and carbon-, water- and soilrelated ecosystem services and noted that monitoring can be done directly,
indirectly or in combination with ecosystem models. Specific efforts to map
and monitor urban ES based on or in connection with remote sensing data
are beginning to emerge (Lakes and Kim 2012; Raciti et al. 2014).
When used in conjunction with remote sensing data and especially Landsat,
ES indicators are often linked to LULC types as the main tool for
measurement of their status. Of most relevance to this research is the use of
land cover as a proxy measure of ecosystem services thanks to its multiple
linkages to a range of services such as watershed protection and carbon
storage (Konarska et al. 2002). Most often, the services provided by the
ecosystem in each land cover type are identified and assigned a monetary
value based on calculations and previous research and these are summed
based on the extent of the particular LULC (Ayanu et al. 2012; and e.g.,
Costanza et al. 1997; Wu et al. 2013). In this way, LULC change can
function as an indicator of change in ecosystem value and to some extent
condition. But the question of how to assign value to ES is problematic and
the focus of much research and debate (Costanza and Folke 1997; Ludwig
2000; Turner et al. 2003; Chee 2004; De Groot et al. 2010; Turner et al.
2010). The study and development of remote sensing indicators to monitor

26

the condition of ecosystem services in urban areas is a relatively new but

rapidly developing field of research.

3 Study areas and data description

3.1 Greater Toronto Area
The Greater Toronto Area is the most populous metropolitan area in Canada
and is one of the fastest growing urban areas in North America. The GTAs
population grew by roughly 1.75 million between 1985 and 2005 to reach a
total of 5.5 million. Located on the northwest shore of Lake Ontario with an
area of 7 125 km2, it includes Metropolitan Toronto and four other regional
municipalities (Durham, Halton, Peel and York) with a combined population
of about six million. The Oak Ridges Moraine (ORM), an environmentally
significant and sensitive area that lies north of Toronto, covers an area of 1
900 km2, is the largest glacial remnant in Ontario and acts as a groundwater
recharge/discharge area for approximately 65 watercourses. In addition to its
importance for water quality in the region, the ORM contains most
remaining natural areas in the GTA bio-region including forests, wetlands,
and various plant and animal species, and provides for most of the
recreational opportunities for the GTAs significant population. Yet recent
increase of built-up areas on the ORM has raised serious concern pertaining
to water quality and quantity. Increased water withdrawal and contamination
could impact those living in the GTA, fisheries, wildlife and conservation
(NRC 2007). Future urban growth within the surrounding municipalities will
have significant impact on the important resources that the ORM provides
and on the overall environmental quality of the region.
Landsat 5 TM imagery with 30m spatial resolution was acquired from three
different years for the GTA study: 18 July and 12 August 1985, 30 July and
24 August 1995, and 2 and 25 July 2005. Two scenes (adjacent orbit paths)
were necessary for each year in order to obtain coverage of the whole GTA.
Images from summers acquired on near anniversary dates/months were
selected to provide the maximum differentiation of major land cover classes
and comparable classification results from various years. The major land
cover/use classes in the area are: water, forest, golf courses, agriculture, lowdensity built-up (residential areas), high-density built-up (including roads
and industrial areas), construction sites and parks/grassy fields. The study
area over the GTA is approximately 7 600 km2 and is outlined in yellow in
Figure 2.

27

Figure 2 2005 Landsat TM images and study extent of the Greater Toronto
Area (Red: TM4, Green: TM5, Blue: TM3) shown with regional municipal
boundaries in yellow, the Oak Ridges Moraine outlined in white and TRCA
watersheds (within which environmentally significant areas were designated)
outlined in dark gray.

3.2 Stockholm region and County

Swedens capital, Stockholm, is located on the countrys east coast and
includes a large archipelago extending into the Baltic Sea. Stockholm
municipality covers an area of around 216 km and has over 900 000
inhabitants, while Stockholm County, a much larger region, covers 6 519
km with over 2 million inhabitants (USK 2011). The population of
Stockholm County grew by about 18% between 1986 and 2006. The
planning authorities in the Stockholm region have called this area of Sweden

28

a green big city region (RTK 2002). The citys unique structure with builtup areas along the radial transport network and non-built up green wedges in
between make convenient connections between work and home possible as
well as providing good access to green areas. Greater Stockholms green
wedges that lead from the countryside in towards the more central parts of
the city and the green links between these wedges comprise the framework
of the regions green structure. These green wedges also provide several of
the Stockholm regions essential ecosystem services. The city of Stockholm
won the title of Green Capital in 2010 from the European Union and has a
vested interest in preserving the balance between its green and built-up
spaces. Yet the regions population continues to grow, putting added
pressure on the natural environment and increasing demand for built-up
areas.
Two separate investigations of the greater Stockholm area were performed.
They were each based on different types of remote sensing imagery and
examined different study area extents, which will hereafter be referred to as
Stockholm region (study area for Paper III) and Stockholm County (one of
two study areas in Paper IV). Paper I, which was limited to Stockholm
municipality as study area, functioned as a methodological test-run for Paper
IIIs expanded study area. Papers I and III present the results from the first
SPOT image-based Stockholm investigation. Four scenes of SPOT imagery
with green, red and near-infrared bands over the Stockholm area were
acquired for this study: two on 13 June 1986, one on 5 August 2006 and one
on 4 June 2008. The images were selected from the peak of the vegetation
growth season to maximize the spectral differences between built-up areas
and vegetation and to avoid detection of unreal changes caused by seasonal
differences between years. Two scenes from 2006 were sought but an
appropriate 2006 growing season image over the northern parts of
Stockholm County was simply not available. The scene from 2008 is
therefore used as a substitute. The SPOT imagery from 1986 was from
SPOT 1 with a resolution of 20m, while that of 2006 and 2008 was from
SPOT 5 with a resolution of 10m.
The specific study area in Paper III is based on the extent of the satellite
images and, excluding an edge buffer, includes all or part of the territory of
18 municipalities in Stockholm County (see lefthand image in Figure 3). The
total land area covered by the satellite images is approximately 2 220 km2.
The major land cover classes in the area are low-density residential areas
(LDB), high-density residential built-up areas (HDB), industrial/commercial
areas, forest, open land, forest and open land mixed, and water. The HDB
class is somewhat unique in this study since much of the high density builtup areas in the city center of Stockholm are composed of buildings that
have businesses on the ground floor but residences on the above floors.

29

Therefore the HDB areas classified in the center of Stockholm do include

some commercial areas, which could at the same time be classified as
residential. In contrast, the industrial/commercial class includes for example
industrial park areas as well as shopping centers.

Figure 3 Study areas from Papers I, III and IV: To the left, the Stockholm
County municipalities included in the study in Paper III are outlined and
labeled in yellow with 2006 and 2008 SPOT imagery as backdrop.
Stockholm Municipality (labeled simply as Stockholm) comprises the
study area in Paper I. To the right, the boundary of Stockholm County is
outlined in yellow with 2009 and 2010 Landsat imagery as backdrop. This is
the extent of the study area in Paper IV.
A number of GIS datasets were collected from several national and regional
Swedish sources, namely Lantmteriet (The Swedish National Land Survey),
Regionplanekontoret Stockholms Lns Landsting (RTK) and Storstockholms
Lokaltrafik (SL). These included data on land cover and transportation
networks (Lantmateriet), public transport stops/stations (Lantmteriet and
SL), and large parks and noise-disturbed areas (RTK). This data was used in
the construction of environmental indicators for the greater Stockholm area.

30

Stockholm County, which covers approximately 7 150 km2 (including

water), is the basis for the study area of the second Stockholm investigation
described in Paper IV (see the righthand image in Figure 3). Six scenes of
Landsat 5 and 7 imagery, two for each decade to cover the entire Stockholm
County area, were acquired: two on 7 July 1989, and one each on 24
September 2000, 4 August 2002, 28 June 2009 and 24 June 2010. In some
cases, the difference in image dates is not exactly 10 years but can deviate up
to two years due to the fact that there are no images available at the same
anniversary or that images that lie closer to the decennial anniversary suffer
from high cloud cover. The images are however acquired in the same
vegetation period from May to September and are considered the most
suitable Landsat images for the purpose of the study. The major land cover
classes in the area, given the type of imagery and classification technique
employed in this study, are low-density residential areas (LDB), high density
built-up areas (HDB) including industrial/commercial areas, forest,
agricultural/open land, parks/urban green areas, and water.

3.3 Shanghai
Shanghai is located on the eastern seabord of China and is a major financial
center for the country. Greater Shanghai is also a productive agricultural area
(Zhang et al. 2011). Population growth in the region has been tremendous
over the past few decades, with an increase of 72% between 1990 and 2010.
From 23.03 million in 2010, its total population is expected to reach 28.4
million by 2025 (United Nations 2012). The rapid urbanization that has
accompanied this impressive growth has put pressure, specifically in the
form of pollution, on the surrounding ecosystem (Ren et al. 2003).
One of the two study areas in Paper IV is the total area of Shanghai, which
measures about 6 340 km2. Land use types in this region include HDB areas,
commercial and industrial areas, ports, airports and residential areas.
Agriculture tends to border the urban areas, with strips of rural areas
including villages and farms. There are varying forms of water such as sea,
lakes, rivers, aquacultures and wetlands. Trees and forests tend to be scarce
and connected stands are found almost exclusively in managed urban parks.
Landsat was chosen as image data for the joint study over Stockholm and
Shanghai given its standardized high quality and immediate availability.
Here again six scenes of Landsat 5 and 7 imagery, two for each decade to
cover the entire Shanghai area, were acquired on the following dates: 18
May 1987, 11 August 1989, 14 June 2000, 3 July 2001, and two on 17 July
2009.

31

4 Methodology
Each study presented here was composed of several methodological steps
ranging from pre-processing of satellite imagery to calculation of
environmental indicators based on classification results. The general
approach in each research was to process the imagery, classify it, assess
classification accuracy, calculate environmental indicators and indices and
evaluate the results. Figure 4 illustrates how the image processing and
classification methodologies compare for each study area. The flowchart in
Figure 5 compares environmental indicator calculation for each study area.

4.1 Image processing

4.1.1 Image pre-processing
Satellite images for the GTA and Stockholm region studies (Papers I-III)
were geometrically corrected with the help of regional topographic maps
based on at least eight GCPs for each image and with a root-mean-square
error of one third of a pixel or lower. The Stockholm/Shanghai Landsat
imagery did not require geometric correction since it was issued by GLS in
the most appropriate coordinate system for the study. Each pair of scenes
was then mosaicked together using neighbourhood color balancing.

4.1.2 Texture analysis with Grey-Level-Co-occurrenceMatrix features

Texture, or the spatial distribution of tonal variations in an image, can aid in
the identification of objects or regions of interest (Haralick et al. 1973).
GLCM-generated texture features have been particularly useful in urban
areas (Shaban and Dikshit 2001; Herold et al. 2003; Ban and Wu, 2005;
Gamba and Aldrighi 2012). GLCM texture analysis was performed in each
study to provide more detailed input to the classification process and to
improve accuracy of the results. Based on trials and findings from past
research (Baraldi and Parmiggiani 1995; Clausi 2002), mean, standard
deviation (variance) and correlation textures of the infrared band for the
GTA and the red band for the Stockholm region study (Papers I and III)
were selected. For the Stockholm-Shanghai study (Paper IV), variance was
calculated on Landsat bands 4 and 5.

32

4.2 Image classification

The classification techniques for the different study areas differed
significantly from this point on, as outlined below.

4.2.1 Maximum Likelihood Classification

The maximum likelihood algorithm (MLC) was used to classify Landsat
imagery based TM bands 3, 4 and 5 and the abovementioned texture features
for the GTA (Paper II). Eight land cover classes were classified: water,
forest, parks, golf courses, agriculture, HDB, LDB and construction sites.
Eight different types of agriculture were initially classified and later
aggregated to form the agriculture class. When issues of separability
between certain urban vs. natural land use types arose, it was decided to
create two broad landuse masks for rural and for urban areas and to perform
two separate MLC classifications under these.

Figure 4 Flowchart comparing image processing and classification

methodologies for each study area

33

4.2.2 Object-based Image Analysis and rule-based

classification
For the Stockholm study utilizing SPOT (Papers I and III), the images were
first segmented using eCognition based on several input data: spectral data,
GLCM texture measures and a preliminary MLC of the SPOT data. The
scale parameter was 20 and the homogeneity criteria were set at 0.1 for
shape and 0.5 for compactness. These parameters were selected based on
trials and provided the most appropriate objects for classification in that they
most closely represented discrete areas of the different land cover types
found in the region, namely water, LDB, HDB, industrial/commercial areas,
forest, forest and open land mixed, and open land. The resulting objects were
then classified sequentially into the major land use types through the
construction of rules based on the objects mean value of various input data.

4.2.3 Support Vector Machines classification

For the Stockholm-Shanghai study utilizing Landsat data (Paper IV), an
SVM classifier was used on bands 3, 4 and 5 plus the texture features
mentioned above to identify the following land cover classes in Stockholm:
LDB, HDB (including industrial/commercial areas), forest, agricultural/open
land, parks/urban green areas, and water. Urban and rural masks were
subsequently used to correct the more consistent misclassifications, most
often caused, in the case of Stockholm, by confusion between HDB and bare
agricultural fields and between forest and LDB areas.
All classifications were filtered as a final processing step.

4.3 Accuracy assessment

Once the classifications were completed, accuracy assessments were
performed using random sample vector points for verification of each land
cover/use category. Overall accuracy, Kappa coefficient, and users and
producers accuracies were calculated and used as accuracy measures.

34

Figure 5 Flowchart comparing environmental indicator calculations for

each study area

4.4 Landscape metrics

Based on the issues outlined in section 2.2.1 related to the use of landscape
metrics, a tailored core set of landscape metrics was selected for analyzing
changes in each of the study locations. Table 1 provides a brief description
of each of the metrics calculated in this research. Table 2 shows in which
study area each metric was applied.
Table 1 Compiled set of landscape metrics used in the current research
Landscape Composition
Metrics
Class Area Percentage
(CAP)
Patch Density (PD)
Mean Patch Size (MPS)
or Area-weighted Mean
Patch Size (AMPS)
Largest Patch Index (LPI)

Description
The percentage or proportion of each class in
the landscape. The Class Area (CA)
calculated by Fragstats is divided by the total
landscape area to obtain this percentage.
Number of patches per square kilometer
(Area-weighted) average patch size of a class
of patches
A simple measure of dominance; quantifies
at the class level the percent of landscape
area comprised by the largest patch.

35

Landscape
Configuration Metrics
Mean Patch Shape Index
(PSI_MN) or Areaweighted Mean Shape
Index (AWMSI)
Area-weighted mean
perimeter to area ratio
(AWMPAR)
Total Edge Contrast Index
(TECI)
Contrast-weighted Edge
Density (CWED)

Area-weighted Mean
Proximity Index (AMPI)
Area-weighted Mean
Similarity Index (AMSI)

Cohesion

Connectance index
(CONNECT)

Contagion (CONTAG)

Description
(Area-weighted) average of the ratio of
perimeter to minimum possible perimeter
given the number of cells in the patch;
measure of shape complexity.
The sum of the ratios of patch perimeter to
area multiplied by proportional patch
abundance
Quantifies edge contrast (degree of contrast
between a patch and its neighbors) as a
percentage of maximum possible
Measured in meters per hectare: CWED = 0
when there is no class edge in the landscape.
CWED increases as the amount of class edge
in the landscape increases and/or as the
contrast along the class edges increases.
Measures relative area-weighted distance
between patches of the same class
measures degree of isolation of
corresponding patch type
Also measures area-weighted distance but
considers size, proximity and similarity of all
patches within a specified search radius of
the focal patch
Measures the physical connectedness of the
corresponding patch type; cohesion increases
as the class becomes more clumped or
aggregated in its distribution.
Measures connectivity of the land cover type
and is defined based on the number of
functional joinings between patches of the
corresponding patch type, where each pair of
patches is either connected or not based on a
user-specified distance criterion. Reported as
percentage of the maximum possible
connectance given the number of patches in
the land cover class
A measure of the relative aggregation of
different patch types at the landscape level:
Contagion approaches 0 when the patch types
are maximally disaggregated and

36

interspersed. CONTAG = 100 when all patch

types are maximally aggregated; i.e., when
the landscape consists of a single patch.
For further discussion of the rationale for selecting the metrics for the GTA
landscape please refer to the Methodology section of Paper II.
For the Stockholm research utilizing SPOT data, the difference in resolution
of the images could compromise the comparability of metrics generated
from them. While the creation of objects helps to reduce this risk, the seven
original land cover classes were further aggregated to three larger classes:
urban, natural and water classes for generation of the metrics. The areaweighted shape metrics are used since they give more weight to larger
patches. The influence of the original difference in resolution of the
classifications might thereby be further minimized since larger patches are
more likely to be agglomerations of objects rather than individuals.
In regard to the metrics used in the Stockholm/Shanghai study (Paper IV),
the area-weighted mean metrics are used rather than their simple mean
equivalents since they provide a landscape-centric perspective of landscape
structure (they reflect the average conditions of a pixel chosen at random).
This landscape-centric perspective is best suited to this research since two
different landscapes are being studied and compared. The contrast-weighted
edge density index (CWED) is used for similar reasons (as opposed to using
the total edge contrast index): the CWED standardizes edge to a per unit area
basis that facilitates comparison between landscapes of different size. Edge
is quantified from the perspective of its functional significance and thus
landscapes with the same CWED would be presumed to have the same total
magnitude of edge effects.
Table 2 Table indicating in which studies each LM was used
Landscape Metric
Class Area Percentage/Proportion (CAP) or
Percentage of Landscape (PLAND)
Patch Density (PD)

Used in
All studies
All studies

Mean Patch Size (MPS) or Area-weighted

Mean Patch Size (AMPS)
Largest Patch Index (LPI)

GTA
Stockholm/Shanghai
Stockholm/Shanghai

37

Mean Patch Shape Index (PSI_MN) or Areaweighted mean shape index (AWMSI)

All studies

Area-weighted mean perimeter to area ratio

(AWMPAR)

Stockholm region

Total Edge Contrast Index (TECI)

GTA

Contrast-weighted Edge Density (CWED)

Stockholm/Shanghai

Area-weighted Mean Proximity Index

(AMPI)

GTA

Area-weighted Mean Similarity Index

(AMSI)

GTA

Cohesion

Stockholm/Shanghai

Connectance index (CONNECT)

Stockholm region
GTA
Stockholm/Shanghai

Contagion (CONTAG)

4.5 Urban and environmental indicators and

indices
4.5.1 GTA
In addition to landscape metrics, other indicators were also explored as a
measure of spatial patterns of urban growth. In particular, indicators
measuring urban compactness that are not dependent on patch shapes were
considered useful. Three of the indicators used were developed by Kasanko
et al. (2007a; 2007b), namely average weighted distance, co-central ring
density and degree of settlements dispersion. All five of the indicators
calculated are presented in Table 3. For details of how these indicators were
selected, used and calculated please refer to the Methodology section of
Paper II.

38

Table 3 Urban indicators used in the GTA study

Urban Indicators
Average Weighted Distance Indicator
(AWDI)
Co-central Ring Density Analysis
(CRDA)
Degree of Settlements Dispersion
(DSD)
Urban Overlap with Environmentally
Significant Areas (ESAs)
Urban Perimeter of Environmentally
Significant Areas

Description
Measures the mean weighted
distance of residential area patches
from the city center
Proportions of urban versus nonurban areas in 5 km rings spreading
out from the city center are
measured
Border length of built-up areas in
each municipality is divided by the
minimum possible length given the
surface area
Degree of overlap of expanding
urban areas with ESAs is calculated
Increase in urban perimeter of ESAs
between 1985 and 2005 is
calculated

Environmentally significant areas (ESAs) were established by the

Metropolitan Toronto and Region Conservation Authority. As discussed in
the Methodology section of Paper II, the change in composition of the
surrounding land use of an ESA determines if it is subjected to either
beneficial or adverse ecological effects.

4.5.2 Stockholm region

Bearing in mind Stockholms regional environmental goals, a series of
environmental indicators were developed and calculated to see how the area
has changed in this regard in recent years. Rankings for each indicator were
later compiled to construct an environmental impact index. The indicators
are described in Table 4 below. For details on the selection process for these
indicators and further explanations, please see section 3.3.2 of Paper III.

39

Table 4 Enviornmental impact indicator specifications used in Paper III

(the unit for comparison is the large forested area or LFA)
Environmental
indicator

Amount of
green areas
important for
ecosystem
services

Definition/
description
For larger forested
areas (min 3 km2),
minimum width
should be 500m
(RTK 2008). All
meet the criteria of
min. 500m width.
Ranked according to
their area (larger is
better).

Calculation method

Unit

Use Zonal Thickness

and Calculate area in
ArcMap.

Square
kilometers

No unit
(index lower is
better)

No unit
(index higher is
better,
represen
ts an
LFA
which is
not as
stressed)

Condition of
green areas
important for
ecosystem
services

Degree of intactness
of forested areas
(how much have the
forests been
disturbed? Clear cut
areas, roads, etc.)

Use perimeter as
indication of perforation
in green areas
(calculated with both
external and internal
boundaries). Compare
this to min. poss. length
given area. P or Cmin =
2Pi*sqrt(A/Pi)
Iperf = Preal/Pmin.poss
(C = circumference, P =
perimeter, A = area, Pi =
3.14...)

Promixity of
green areas
to intense
urban land
use

Area Weighted
Distance between
green areas defined
above and all HDB
and industrial/
commerical areas,
with size of the LFA
and urban areas
taken into account.
The nearness of
dense urban and
industrial areas
places pressure on
green areas. This

Iprox = ALFA *
(D2/ADU)
(A = area, D= distance,
DU = dense urban
patch)
Use centroids of LFAs
and HDB/IC areas. Use
"Point distance" in
ArcMap.

40

index is intended as a
rough measure of
that pressure.
Degree of
urban
perimeter
surrounding
LFAs

Urban areas in direct

proximity to forested
areas place pressure
on these, through
pollution and cutting
off LFAs from other
natural areas, for
example.

Noise level

To what extent are

LFAs affected by
noise pollution?

Create layer of urban,

water, open and
vegetation LC
categories. Combine this
layer with LFA line file
using "Identity" to find
degree of edge with
urban areas.
Use RTKs Stockholm
county 40-65 dB dataset
(buffered road/rail
network) created 2004.
Use "Identity" on LFA
layer with noise layer
and export the over 55
dB overlap areas. Area
in meters of LFA
affected by highest
noise levels (55-65). 55
dB is WHO's outdoor
noise limit (WHO
1999).

Percent
urban
perimeter

Area of
LFA
affected
by >55
dB (km)

With the indicator results as input, the compilation of the Environmental

Impact Index is intended to assess broader change in the natural environment
and the ecosystem services it provides to the region. The method for
compiling the index is simple with no weighting system in order to ensure as
much transparency as possible. Once all indicators were calculated, the
analysis units were ranked according to each indicator. An average of the
indicator rankings was taken to provide an overall index score. The major
advantage with this index and these indicators is that they are comparative
(the same methods and standards are used to derive them from each
classification for each analysis unit). They can therefore be used as a
measure of the differences between 1986 and 2006.

4.5.3 Stockholm and Shanghai comparison

In order to compare the speed and magnitude of urbanization, two indices
are calculated for all three decades in both locations in the

41

Stockholm/Shanghai study. The Urban Land Index is the ratio of urban land
to total land and the Urban Expansion Index is the difference in urban land
over a period of time as calculated by Hu et al. (2009). An Urban Green
Index was also devised to quantify the development of urban green spaces
(UGS) in comparison to simultaneous urban development. The index is
calculated as the ratio of UGS increase divided by the sum of increases in
HDB and LDB. For details on the calculation of these indices, please refer to
section 3.5 of Paper IV.

4.6 Valuation of Ecosystem Services

Ecosystem service value for each decade in the Stockholm County/Shanghai
study was calculated based on the valuation scheme of Costanza et al.
(1997). The scheme is reported to account for the following ecosystem
services: gas regulation, climate regulation, disturbance regulation, water
regulation, water supply, erosion control, soil formation, nutrient cycling,
waste treatment, pollination, biological control, habitat/refugia, food
production, raw materials, genetic resources and recreational and cultural
services. Table 2 under Section 3.7 of Paper IV lists the land cover classes
identified in the study area, their corresponding biomes and ecosystem
service values per hectare and year based on Costanza et al. (1997).
It should be noted that in Costanzas original valuation there was neither a
value attributed to urban areas nor for UGS. Since UGS such as parks or golf
courses are predominately composed of trees, water bodies and lawns, a
value has been defined in this study by averaging the values of forests,
grasslands and lakes/rivers and assigned to UGS.

5 Results and discussion

5.1 Image classifications
The accuracy results for each of the classifications performed in this research
are reported in Table 5. Relatively high accuracies were obtained in all
cases. MLC classification under urban and rural masks achieved the highest
accuracies, followed by segmentation and rule-based classification and then
SVM classification. It is perhaps a little surprising that the MLC approach
provided the best results given that OBIA and SVM are newer approaches
and have often been cited as better alternatives to MLC in the literature.
Quality of the input imagery is of course a factor since noise-free and

42

normally distributed images are easier to classify. The quantity and quality
of the training data are also important for achieving high accuracy
classification results, even more so than the choice of algorithm (Li et al.
2014). The classes to be defined for each region also varied slightly: golf
courses and parks each had their own class in the GTA whereas UGS
covered both of these in the Stockholm/Shanghai study. The GTA also had a
class for construction sites and for the Stockholm region industrial areas
were distinguished from HDB. For more details on the respective
classification accuracy results, see section 4b of Paper I, the Classification
section under Results and Discussion in Paper II and sections 4.2 of Papers
III and IV.
Table 5 Comparison of overall classification accuracies and kappa
coefficients for all classifications performed
Classification type,
Overall accuracy in
location and year
percent
MLC with urban/rural masks:
GTA 1985
94.9
GTA 1995
92.9
GTA 2005
94.3
Object- and rule-based classification:
Stockholm region 1986
92.4
Stockholm region 2006
92.1
SVM:
Stockholm County 1989
90.0
Stockholm County 2000
89.0
Stockholm County 2010
88.2
Shanghai 1990
88.1
Shanghai 2000
87.8
Shanghai 2010
89.4

Kappa coefficient
0.94
0.92
0.94
0.91
0.91
0.88
0.87
0.86
0.86
0.86
0.88

For the GTA classification, the use of the agricultural and urban masks
significantly improved the classification of agriculture and parks for all three
decades, as well as construction sites vs. bare soil. See the Results and
Discussion section of Paper II for comparison with the original
classifications (without urban/agricultural masks). The landcover
classifications for the GTA in 1985 and 2005 are shown in Figure 6.

43

Figure 6 Land cover classification results for the GTA in 1985 and 2005

44

The landcover classification results for the Stockholm region are shown in
Figure 7. In regard to the lower resolution of the 1986 SPOT imagery over
the greater Stockholm area, the objects generated by the segmentation for
1986 were for the most part of a similar size as those for 2006, but the 2006
segmentation did have a greater amount of smaller objects which led to a
slightly more detailed classification, especially when it came to correctly
isolating forest and open land mixed, and to identifying very small green
areas (trees, grass, parks, etc.) within LDB areas.

Figure 7 Land cover classification results for Stockholm region in 1986 and
2006
The landcover classification results for Stockholm County and Shanghai are
shown in Figure 8. For Stockholm County, there was a slight over-detection
of LDB and UGS in the 2000 classification as well as an under-detection of
HDB areas. This was mainly attributable to noise in the 2000 image, which
dampened the spectral differences between these areas. Agricultural areas
were slightly under-detected in 1989 and 2010 and forested areas slightly
over-detected for these years due to a certain type of bright vegetation that
was confused with forest. UGS was also consistently difficult to classify
because of its spectral similarity to agricultural and LDB areas. The use of
rural and urban masks helped to improve the classification of UGS
somewhat.

45

In all the studies, the texture features were instrumental in achieving

sufficiently high accuracy in the classification results. In the GTA, this was
especially true in helping the maximum likelihood classifier to distinguish
land cover types such as golf courses from parks/pasture and agricultural
areas. For Stockholm, texture was indispensable when it came to isolating
built-up areas such as LDB from other types of mixed but natural landscape
and HDB from industrial areas.

Figure 8 Land cover classification results for Shanghai (top row) and
Stockholm County (bottom row) in (from left to right) 1989/90, 2000 and
2010

46

5.2 Landscape change and potential

environmental impact analysis using
landscape metrics
5.2.1 Urban growth and landscape change trends
Figure 9 illustrates the growth in urban areas and decrease in natural and
semi-natural land cover in each of the study areas during the years examined.
Urban land cover in this instance includes HDB, LDB and industry or
construction sites when applicable. All other LC classes are included in the
non-urban category with the exception of water, which is excluded entirely.
The figure indicates relative rates of growth/loss and it becomes clear that
Shanghai is urbanizing and losing non-urban areas most rapidly, followed by
the GTA at a much more moderate rate and finally Stockholm with the
lowest rate. In addition, the growth and loss rates themselves in Shanghai are
increasing while those for the GTA and Stockholm are relatively steady.
What also emerges clearly from this chart is the fact that the GTA and
Stockholm are landscapes still clearly dominated by more natural landcover
types. For Shanghai, however, it appears that a shift in dominance from more
natural to urban LC types is imminent and could perhaps have happened
already given the trajectory displayed. There is still a large difference in
relative amonts of urban vs. non-urban land cover types in the GTA and
Stockholm, which indicates that a shift in landscape dominance is not
imminent. It appears that these landscapes will remain dominated by more
natural LC types for some time to come. More specifically, according to the
largest patch index results at the class level, the GTA is dominated by
agricultural land and Stockholm County by forest. For Shanghai, a transition
occurred from dominance by agricultural land before 2000 to HDB
afterwards.

47

6000

Square kilometers

5000

GTA urban
GTA non-urban

4000

Shanghai urban
Shanghai non-urban
Stockholm Co. urban

3000

Stockholm Co. non-urban

Stockholm reg. urban

2000

Stockholm reg. non-urban

1000
0
1980

1985

1990

1995

2000

2005

2010

2015

Year

Figure 9 Growth trends for urban vs. non-urban land cover in the GTA,
Shanghai, Stockholm County and region during the period 1985-2010
Figure 10 shows the trend of contagion values for the GTA, Stockholm
County and Shanghai. All decreased signifying that each landscape was
becoming more disaggregated and patch types were more interspersed. Here
again the different rates at which this was occurring are of interest. The GTA
landscape shows the fastest, steadiest and most dramatic shift towards
dissaggregation. Shanghais trend picks up speed after the year 2000 while
the rate for Stockholm County slows down following that year. Indeed the
rates for these two landscapes appeared to be very similar to begin with but
later clearly diverged, with Shanghais rate nearing that of the GTA in the
second decade. The information contained in this chart points generally to
how quickly these landscapes are becoming fragmented. More specific
information about landscape change results in each of the study areas is
reported in the following sections.

48

64

Contagion value

62
60

GTA
Stockholm Co.

58

Shanghai
56
54
52
50
1980

1985

1990

1995

2000

2005

2010

2015

Year

Figure 10 Contagion trends for the GTA, Stockholm County and Shanghai
during the period 1985-2010

5.2.2 GTA
Urban areas in the GTA grew by about 39% between 1985 and 2005. With
regard to the landscape composition metrics, the most significant
proportional results in the GTA as a whole are in regard to loss of
agricultural lands and parks in favor of low-density built-up areas and golf
courses. Forested areas remained relatively unchanged, comprising around
one-fifth of the landscape over the twenty year period. Only agricultural
areas decreased, while nearly all other categories increased over the same
time period, most significantly LDB areas and golf courses. The most
relevant LM results are presented graphically in the Landscape Metrics
section under Results and Discussion in Paper II.
It is positive that the amount and size of forested areas remained fairly
constant over the two decades since forest constitutes the most natural
habitat available in this landscape and is crucial to its ecological function.
But the trend of smaller and fewer patches of agriculture and parks/grassy
fields (the two semi-natural land cover types in the landscape) may have a
negative effect on species diversity and persistence. Moreover the trend of
low-density built-up areas replacing agricultural areas can directly impact
species populations in relation to connectivity. So while it is positive that
forest areas remain relatively unchanged, isolation of this important habitat
increased due to the replacement of agricultural patches with low-density

49

built-up areas. This occurrence may hinder how natural processes operate
across the whole landscape.
The landscape configuration metric results confirm increased fragmentation
of the landscape due to urban development and generally indicated increased
vulnerability for natural and semi-natural land cover types and the
biodiversity that depends on them through, for example, greater edge,
contrast and lowered connectivity. For more detailed discussion of the LM
results and what they mean for the GTA landscape and the ORM in
particular, please see the Landscape Metrics section under Results and
Discussion in Paper II.

5.2.3 Stockholm city and region

Results of the landscape metrics analysis for Stockholm city reveal that
overall non-built up areas decreased by approximately 1.7% while built-up
areas increased by around 2.1% over the twenty year interval. The HDB,
industrial, and mixed forest and open land classes increased at the expense of
open land, forest and LDB areas, the most significant among these being the
increase in HDB (over 2%) and the decrease in open land (~1.5%), see
Figure 2 in Paper I. Regional authorities have clearly expressed the aim of
building more dense urban areas rather than taking from the more natural
surrounding landscape for urban growth (RTK 2002). While there has
evidently been some loss of non-built up land, the goal of building more
densely is apparently to some extent being realized within the city of
Stockholm. See section 4c of Paper I for more details.
But results for the greater Stockholm region paint a slightly different picture.
Looking at the change in class area percentage between 1986 and 2006,
natural land cover dropped and urban areas increased by approximately 1.5%
of the whole landscape or just over 33 km2. Urban areas increased by just
under 10% while natural areas decreased by 2%. Calculated LM results
indicate that both became more fragmented: natural areas became more
isolated or shrank whereas new small urban patches came into being.
Sections 4.3.1 and 4.3.2 of Paper III provide more information on landcover
changes that occurred during this period in the greater Stockholm region.
The LM results reveal that urban growth in the greater Stockholm region
occurred mainly at the expense of natural land cover types while this was not
entirely the case for the city of Stockholm during the same period (where
there was some conversion of LDB to HDB). This circumstance is not
necessarily surprising, since the natural areas within the city are limited and

50

often protected, while those outside are more plentiful and frequently carry
fewer restrictions when it comes to development.
Of note is that the difference in resolution in the Stockholm SPOT imagery
limited the use of LM for comparison purposes since it gave rise to an
inconsistent delineation of the forest and open land mixed class between
classifications. This by extension caused problems for comparability of LM
results for the forest, open land, and forest and open land mixed LC classes.
Future research will seek to rectify this situation (see Section 6 below).

5.2.4 Stockholm County and Shanghai

Between 1989 and 2010, the speed and magnitude of urban growth in
Shanghai exceeded that of Stockholm County by far. In total, urban areas in
Shanghai increased by 120% or 1 768 km2. In Stockholm, a 12% increase in
urban areas could be observed, which corresponds to an area of 100 km2.
This increase in both HDB and LDB areas came predominately at the
expense of agricultural land in both study areas. From the landscape metrics
results, we can deduce that the growth of HDB areas was mainly responsible
for increased fragmentation in both of the city regions as new patches were
added in previously more natural areas, while LDB areas tended to expand
through consolidation with pre-existing patches.
Based on the statistics generated from the land use classifications of
Stockholm County, urban areas grew by about 7% or 57 km2 between 1989
and 2000 and by about 5% or 43 km2 between 2000 and 2010 in Stockholm
County. The percent of urban growth between 1989 and 2010 was over 12%
or 100 km2. This is relatively similar to the 10% increase found at the
regional level. Looking at the change in class area percentage between 1989
and 2010, natural land cover dropped and urban areas increased by
approximately 1.4% of the whole landscape (just over 100 km2), which is
also very similar to the change for regional Stockholm (1.5%). With regard
to the landscape composition metrics, the most significant proportional
changes are in regard to loss of agricultural/open land in favour of lowdensity and high density built-up areas. Forest remained relatively
unchanged during the two decades.
The LM results indicate both positive and negative potential impacts for the
regional ecosystem in Stockholm. On the one hand, it is certainly positive
that the amount and size of forested areas has remained constant over the
two decades and that forest dominates the landscape through by far the
largest patch; forest constitutes the most natural habitat available in this

51

landscape and is crucial to its ecological function - for air filtration,

rainwater drainage, microclimate regulation and noise reduction (Bolund and
Hunhammar 1999). On the other hand, the trend of smaller and fewer
patches of agriculture (a semi-natural land cover type in the landscape) may
have a negative effect on species diversity, movement and persistence. Both
the patches and the species that depend on them become more vulnerable.
The configuration metrics revealed a greater amount of edge interface
between forest and urban areas in Stockholm. The increased interface
between natural land cover types and urban areas tends to have a negative
impact on the flora and fauna found in the vicinity as it increases their
vulnerability to adverse external effects. In addition, the agricultural class
lost some physical connectedness and this could mean loss of more natural
LC conduits for fauna moving between forested areas.
In terms of natural LC important for the regional ecosystem in Shanghai,
there was some loss of the already relatively small amount of forest left in
the region as well as wetlands. As seen in Figure 5.5, the contagion metric
for Shanghai continuously decreased over the two decades indicating a
transition towards a more fragmented landscape. From an ecological
perspective, a higher degree of fragmentation, especially through built-up
areas and a lower degree of patch connectivity, has negative effects upon
flora and fauna. The amount of natural habitat is reduced and the ecological
quality of habitat edges decreases through often human-induced external
effects, such as carbon dioxide emissions along roads or contaminant loads
from industry or noise and light pollution. Dispersal capacities of species
may be limited due to fragmented green spaces and the genetic pool is
decreased through the inhibition of genetic interchange due to infrastructural
boundaries. For more detailed discussion and graphic representation of the
LM results in both Shanghai and Stockholm County, please refer to section
4.4 of Paper IV.

5.3 Assessing impact with urban form,

environmental and ecosystem service
indicators
5.3.1 GTA
Results from the urban form indicators showed how urban areas changed
geographically in the GTA. The average weighted distance dropped for both
decades, revealing that much of the urban growth occurred relatively close to
the urban center of the region, that additional urban areas were being added
close to or directly amongst previously existing urban areas. The co-central

52

ring density analysis indicated that urban proportions increased in nearly all
areas outside of the metropolitan center over the two decades, including on
portions of the ORM. Dispersion of settlements increased on the ORM
between 1985 and 1995, while for the GTA and on a municipal average the
more significant increase occurred between 1995 and 2005. The majority of
ESAs were increasingly surrounded by urban areas between 1985 and 2005,
furthering their isolation from other natural areas and exposure to negative
impacts from adjacent development. For more details and graphic
representations of these results, see the Urban Compactness Indicator and
ESA sub-sections under Results and Discussion in Paper II.

5.3.2 Stockholm region

For the study over the Stockholm region, the environmental impact indicator
results consist of a series of tables that can be found in the Appendix.
Positive changes in area and perforation signified regrowth of forest in that
particular LFA. Changes in rank for the proximity to intense urban land use
indicator depended mainly on the location of new industrial/commercial or
HDB areas. Negative changes in the percent of urban perimeter indicator
were caused by construction of new LDB areas or roads in direct proximity
to LFAs. Positive changes in the area affected by noise were most often due
to loss of part of the area in the LFA affected by noise. Each LFA is ranked
according to each indicator and a subsequent average ranking across the
indicators provides its overall EI index score. These overall index scores for
both years are shown in Table 5 and Figure 3 in section 4.3.3 of Paper III.
The most noticeable changes in terms of environmental impact and urban
expansion were in the east and north of the study area. Although a good deal
of the urban expansion between 1986 and 2006 occurred in the northwest,
the LFAs in that area were relatively small. There are larger and more far
reaching LFAs in the northeast and these registered more of the effects of the
urban development.
The results from the environmental impact index showed that the best
average ranking scores occur in the northern and eastern portions of the
study area. The worst tended to be closer to the center of the Stockholm
region with a few exceptions. When it came to negative changes in rankings,
areas in the northeast dropped the most. This shift in the northeast was
mainly due to construction of new LDB areas. Generally, in terms of
changes in rankings between 1986 and 2006, the most improved analysis
units are those that are closest to the central Stockholm area, while the least
improved or worsened units tend to be on the outskirts of the study area due

53

to the fact that the urban expansion occurred mostly in the suburban areas.
For graphic depictions of the index, more specific information about the
changes and discussion of their causes, see section 4.3.3 of Paper III.

5.3.3 Stockholm County and Shanghai comparison

Table 5 under section 4.3 of Paper IV shows the urban land, urban expansion
and urban green index results for Shanghai and Stockholm County between
1989 and 2010. Urban green spaces in Stockholm grew by about 11% and in
Shanghai by 425% over two decades. The total increase in built-up areas in
Shanghai was about 120% from 1989 to 2009. Urban expansion proceeded
slightly faster in the second decade than in the first. Urban growth in
Stockholm is also apparent but at a much slower pace, especially from 2000
to 2009. Urban expansion both in terms of speed and spatial extent occurred
predominately from 1989 to 2000. A modest increase of urban areas of 1.5%
can be observed over two decades in Stockholm, which corresponds to an
expansion of about 12% of their original extent.
Tables 7-9 under section 4.5 of Paper IV provide information on how ES
values changed over the two decades in both Shanghai and Stockholm.
While ES values remained rather steady and increased slightly in Stockholm,
LULC changes in Shanghais metropolitan area resulted in drastic ES value
losses. The modest expansion in urban areas that occurred in Stockholm was
accompanied by an increase in urban green spaces (mainly golf courses) that
may enhance quality of life for Stockholm residents and the regions
attractiveness to visitors. The growth in urban green spaces actually
increased ES due to the higher ecosystem service values attributed to urban
green spaces (compared to those for agricultural land which was replaced).
In 1989, ecosystem service values were almost equally high in both study
areas. While the value in Stockholm slightly increased, a substantial loss in
Shanghai was observed reducing the value of ES to 63% of their initial
value.
The ecosystem service valuation results showed a large loss of ecosystem
service clearly for Shanghai, but the results for Stockholm were less clear.
This tool may work relatively well for cities like Shanghai where urban
growth and landscape changes are extreme, but it is more problematic when
applied to regions like Stockholm whose urbanization and landscape changes
over decades are much subtler and where ecologically important urban
features for the city such as allotment gardens (Barthel and Elmqvist 2007)
cannot be integrated into present valuation approaches. The current valuation
approach is not refined enough to capture smaller yet important and perhaps

54

cumulative losses in terms of ecosystem service function in more moderately

urbanizing areas.

5.4 Comparison of the GTA, Stockholm and

Shanghai investigations
While it may be premature to draw definitive conclusions in comparing the
methods and results of the studies conducted in this research, certain issues
and ideas have already become clearer with hindsight. Each study has of
course certain advantages and drawbacks. Observation at three different
points in time, instead of only two, is valuable since this third point helps to
establish and evaluate a trend. It is also of course much better to use the
same data resolution for the points in time being compared; it makes use of
landscape metrics more reliable, simplifies the classification process and
makes the classifications more directly comparable. One clear advantage
with performing multiple Stockholm studies is the different scale
perspectives that are thereby provided. Many researchers have emphasized
the importance of scale when it comes to environmental change (Bissonette
1997; Gibson et al. 2000; Wu 2004). A more thorough look at the
differences in the results between these different scale studies over
Stockholm and what they imply should be part of future research.
Is it feasible to make a direct comparison of the urban growth rates and
patterns in these three very different regions given the large difference in
their population sizes? The land areas of each of the study extents are
relatively close in size. The number of people living in the GTA is about
three times the population of Stockholm, while Shanghais population is two
and a half times the population of Sweden. Given this context, it seems that
urban growth rates mirrored to some extent the relative population sizes.
Shanghais urban growth of 120% was tremendous, the GTAs between
1985 and 2005 was impressive at nearly 40%, while that of the Stockholm
region, at just under 10%, and Stockholm County, at 12%, could be safely
termed as modest in comparison. Again, it seems that population sizes could
probably act as predictors of increase in urban areas: just over 100 km2 in
Stockholm County, nearly 600 km2 in the GTA and over 1700 km2 in
Shanghai.
Does this mean that urban areas in the Stockholm region were growing more
sustainably than in the GTA or Shanghai? This question is more difficult to
answer than it would first appear. The layout and geography of the
landscapes themselves may also play a key role in shaping urban growth in
these areas. For example, Stockholms geomorphology is unique and the

55

growth of urban areas is undoubtedly constrained by the many surrounding

bodies of water. The city center itself is essentially a group of islands and
urban expansion there, with its protected green spaces, is extremely limited.
Growth is thereby diverted to the rural-urban fringe, either to the north or
south of the city center, given the inlet from the Baltic Sea to the east of
Stockholm which opens up into Mlaren on its west side. This configuration
may be pushing urban growth in the Stockholm area towards more
suburbanization, which is dominated by single-family residential
development (Seto et al. 2010), than otherwise might have occurred. This is
also why the urban compactness indicators applied to the GTA would not
have produced as meaningful results if applied to the Stockholm regional
landscape.
The GTA also has its own constraints. Toronto is built on the northern shore
of Lake Ontario and, probably given the desirability of living/working in
proximity to water, the city has spread east-west along the shoreline and then
north. This geographical layout also allowed the GTA to have developed a
more compact urban area at its core. The rural-urban fringe in the GTA is
nearly always characterized by an interface between urban and agricultural
areas, while in Stockholm this interface is between urban areas and forested
or open land. Shanghai shares several of the GTAs geographical and
physical characteristics, such as proximity to water, a compact urban core
and an urban-agricultural interface at its rural/urban fringe. Given these
conditions, the GTA and Shanghai may be more prone to peri-urbanization
(Seto et al. 2010). The danger and possible solution for this condition are
noted by Lambin et al. (2001): In the developed world, large-scale urban
agglomerations and extended peri-urban settlements fragment the landscapes
of such large areas that various ecosystem processes are threatened.
Ecosystem fragmentation, however, in peri-urban areas may be offset by
urban-led demands for conservation and recreational land uses. The TRCA
itself appears to be evidence of such an attempt to offset fragmentation in the
GTA. Establishment of a corresponding authority might greatly benefit the
Shanghai region.
The planning authorities of Stockholm County place great emphasis on the
preservation and importance of green spaces, perhaps more so than those of
the GTA who tend to focus on watershed conservation. But the issues are
also more self-contained for the Stockholm region, where there is only one
county organization making policy decisions. Jurisdiction over natural areas
in the GTA is shared by at least two if not more authorities on different
political levels (Toronto and Region Conservation 2014). Neither is
Stockholm dealing with as big an influx of people or growth pressure as the
GTA, not to mention Shanghai. This influx and urban growth may also be
influenced by factors such as rent controls, which are still common in

56

Stockholm and have a two-fold impact: they create high demand for limited
housing and discourage the construction of more. Urban growth is in this
sense freer due to a more competitive market in the GTA (and Shanghai?)
than in Stockholm.
Whatever the urban growth pattern or cause of it, the nature of the
environmental impact was varied in the different study areas. There was
similarity in the location of new urban areas along the outskirts of existing
urban areas, since the center of each city did not leave much room for urban
expansion. Urban expansion in the GTA tended to repossess solely
agricultural land, while open/agricultural land and forest was converted in
Stockholm. Far more land was converted to urban use in the GTA and
Shanghai, which also have much larger urban areas and populations to begin
with. The environmental impact that occurred in the GTA was characterized
by urban expansion onto the ORM and by increased urban surroundings for
nearly all ESAs, in essence furthering their isolation from the larger
ecosystem. Loss of wetlands and already sparse forest comprised the bulk of
negative environmental impact in Shanghai whose ES values dropped
dramatically. Environmental impact in the Stockholm region was mainly
characterized by fragmentation and loss of portions of LFAs due to
construction of roads and expansion of LDB areas. At the county level,
forests appeared to remain more or less intact while mainly open land was
converted (as was the case in the GTA). The choice of environmental
indicators for impact assessment was tailored to the very different
characteristics of these urban areas. Their selection was heavily influenced
by what local planning authorities regarded as important in terms of
environmental quality. The EI indicators were designed based on
environmental information and goals for the Stockholm region, which
rendered their results site-specific and hopefully more useful. To
complement these place-specific analyses, landscape metrics were used to
gain an overview of the spatial-temporal changes in all three areas and are
slightly more comparable, although there were limitations for the Stockholm
region study.
LM were most successfully used to evaluate potential environmental impact
in the GTA study due to the establishment of their connection with regional
environmental conditions and ecological principles as laid out by the TRCA
(see the Methodology and Results sections of Paper II). Estimation of
probable environmental impact in the Stockholm/Shanghai study was
weaker due to the lack of specific regional ecological information and
principles to which the LM results could be tied. LM can be applied to any
landscape with a high quality classification but credible inferences about
potential environmental impact are more difficult to make from their results
unless sufficient general information about region-specific environmental

57

conditions and ecosystem function has been obtained. A widely applicable

remote sensing-based EIA approach appears to be possible up to a point selection of the indicators used within it must be tailored to the site in
question in order to obtain meaningful and useful results. Remote sensing
studies often involve a top-down approach but a bottom-up complement is
necessary when utilizing this kind of data for regional environmental impact
assessment.
A general observation can be made about the research process on
environmental change and the impact of urbanization at this regional level.
Obtaining information about local ecosystem functions and services at work
in the specific regions is crucial to selecting appropriate indicators and
making relevant connections between indicator results and potential
environmental impact. But this can also be challenging as reliable
information is not always available ecological information appeared to be
much more difficult to obtain for Shanghai than for Stockholm, for example.
Then there is the question of the type and usefulness of information that is
available. The TRCA provides good general environmental information for
the portion of the GTA under its jurisdiction but relatively little is known
about the specific ecosystem services that the ORM provides. In Stockholm,
there is a plethora of environmental information available on the municipal
level (for example, biotope maps, habitat maps and recent species citings),
but there is relatively little information available about green areas and
ecosystem function at the county level. Such conditions can limit an
investigation by changing its direction or extent.

6 Conclusions and future research

6.1 Conclusions
This research demonstrates the utility of indicators derived from remote
sensing in assessing both the spatio-temporal dynamics of urban growth and
its environmental impact in several different metropolitan regions. Differing
classification techniques were employed for the unique study areas given
their individual characteristics and the remote sensing data and tools
available for analysis. Landscape metrics calculated based on the
classifications were useful in quantifying the urban growth trends for each
study area, facilitating their comparison, and drawing conclusions about
potential environmental impact when linked to region-specific
environmental information. Environmental indicators were selected based on
the specific environmental characteristics and issues emphasized by local
planning authorities in each region when possible. The indicator and index

58

results highlighted challenges in terms of sustainable urban growth unique to

each landscape, both spatially and temporally.
The research concerning the GTA demonstrates that MLC classification of
multitemporal Landsat TM imagery and their texture measures can be used
to map spatio-temporal patterns of urban expansion with high accuracy with
the help of urban/rural masks. The results show that urban areas in the GTA
grew by about 20% between 1985 and 1995 and by about 15% between 1995
and 2005. Results from the landscape metrics and urban compactness
indicators show that low-density built-up areas increased significantly in the
GTA between 1985 and 2005, mainly at the expense of agricultural areas.
The contagion index values for both the GTA and the ORM decreased over
this period of time, indicating increased fragmentation of these two
landscapes. Yet the AWDI also dropped for both decades, revealing that
much of the urban growth occurred relatively close to the urban center of the
region, that additional urban areas were being added close to or directly
amongst previously existing urban areas. Nevertheless, as shown by the
CRDA, urban proportions increased in nearly all areas outside of the
metropolitan center over the two decades, including on portions of the ORM.
Dispersion of settlements increased on the ORM between 1985 and 1995,
while for the GTA and on a municipal average the more significant increase
occurred between 1995 and 2005. The majority of ESAs were increasingly
surrounded by urban areas between 1985 and 2005, furthering their isolation
from other natural areas. These findings may encourage concerned
organizations and authorities in Ontario to review policy regarding the ORM
and watershed protection in the GTA.
The research involving the greater Stockholm region reveals the
effectiveness of a knowledge- and rule-based OBIA approach to classifying
multitemporal SPOT imagery and texture features to map spatio-temporal
patterns of urban growth with high accuracy. Urban areas in the Stockholm
region increased and rural/natural land cover types decreased by about 1.5%
of the study area or 33 km2. The landscape change metrics indicated that
both became more fragmented: natural areas became more isolated or shrank
whereas new small urban patches came into being. The most noticeable
changes in terms of environmental impact and urban expansion were in the
east and north of the study area. Although a good deal of the urban
expansion between 1986 and 2006 occurred in the northwest, larger and
more far reaching LFAs in the northeast registered more of the effects of the
urban development. The results from the EI index showed that the best
average ranking scores occur in the northern and eastern portions of the
study area, while the worst tended to be closer to the center of the Stockholm
region with a few exceptions. However, areas in the northeast decreased
most in EI ranking between 1986 and 2006, while the most improved

59

analysis units were close to the central Stockholm area. The shift in the
northeast was mainly due to construction of new LDB areas. While these
changes are subtler than those in the GTA, information about them may help
inform future policy choices by Stockholms regional planning authorities.
The Stockholm County/Shanghai comparative study demonstrated that SVM
classified multitemporal Landsat imagery can be used to map spatiotemporal patterns of urban expansion with high accuracy with the aid of
texture measures and urban/rural mask refinements. Urban areas increased
ten times as much in Shanghai as they did in Stockholm, at 120% and 12%
respectively. Between 1989 and 2010, both HDB and LDB areas in
Stockholm County grew mainly at the expense of agricultural areas. The
landscape metrics results show that fragmentation in both study regions
occurred mainly due to the growth of high density built-up areas in
previously more natural environments, while the expansion of low density
built-up areas was for the most part in conjunction with pre-existing patches.
The growth in urban areas resulted in ecosystem service value losses of
approximately 445 million USD in Shanghai, mostly due to the decrease in
natural coastal wetlands, while in Stockholm the value of ecosystem services
changed very little. The Shanghai landscape has undergone swifter urban
growth and a resulting heavier environmental impact on the regional
ecosystem than has Stockholm.
Some general conclusions can be drawn from analysis of the results. Pixelbased classification methods such as MLC and SVM can work well at a
regional scale and are applicable to different landscapes when used in
conjunction with GLCM texture measures and urban/rural masks. Objectand rule-based classification also appears to perform well at this scale with
the use of texture but further research is needed to evaluate temporal and
geographical comparability. LM work well for EIA in different locations as
long as they can be tied to region-specific environmental information,
principles and/or goals. Environmental indicators should whenever possible
be tailored to the site in question in order to provide meaningful results.
Evaluation of change in ESV as applied here seems to work well for
dramatically urbanizing landscapes but is not specific enough for application
to areas undergoing modest urban expansion.

6.2 Future research

Future research is planned to address some research issues related to this
study as already mentioned. One intended avenue of future research would
address problems related to the comparability of the Stockholm region

60

classification results. SPOT-1 pan imagery from 1987 with a resolution of

10m has been acquired over the study area and is intended to be fused with
the 1986 imagery in order to supply a higher resolution image for
classification and thereby ensure that the classification results are more
directly comparable. A wider range of landscape metrics can then be
calculated and a better separation of some of the natural landcover classes,
such as forest and open land mixed, should be possible. An improved 1986
Stockholm classification and the generation of more extensive landscape
metrics as part of future research would enable better comparability of the
urban growth patterns at different scales and possibly between different
locations. A closer examination of the Stockholm studies performed at
different scales might yield interesting insights into the role of scale and how
it affects estimation of environmental impact from urbanization. 1m
resolution data is available in Stockholm and this may offer a third scale for
comparison and a better possibility for connections with local ecological
data.
As was noted in the results section of Paper III discussing LFA
environmental indicators, an indicator weighting system could be developed.
Although a justified system of weights is notoriously difficult to establish,
methods such as the analytic hierarchy process (AHP) could be tested. AHP
provides a proven and effective way of dealing with complex decision
making by organizing and assessing alternatives against a hierarchy of
multifaceted objectives (Saaty 2001).
The Shanghai study provides a look at a case of fast-paced urban growth,
while an example of modest regional urban growth has been established by
two studies over Stockholm. The GTA investigation provides information
about a region urbanizing at a moderate rate. One intention is then to
compare and contrast urban growth patterns and environmental impact
results from these very different regions of the world. The challenge will be
to find valid methods of comparison given the regional differences. It may
be useful to set the results within a framework of urban growth theory in
order to facilitate comparison. An additional focus will be on evaluation of
these and possibly additional classification and indicator methods in terms of
finding a widely applicable approach to estimating environmental impact
from urbanization in metropolitan landscape regions based on optical remote
sensing data.
It will also be important to look further into which indicators could be
helpful in gauging ecosystem service condition at the landscape level and
how to link the landscape matrix with ecosystem services in order to draw
conclusions about environmental impact based on changes in land cover due
to urbanization. Recent research has begun to address the latter issue (Frank

61

et al. 2012; Jacobs et al. 2014). CICES (Haines-Young and Potschin 2013)
and conceptual monitoring frameworks may prove helpful in linking land
cover types to provisioning and regulation ecosystem services. Yet local
environmental information is still indispensable for meaningful linkages and
ecological field data would be a welcome addition to future research as
reference and evaluation information when drawing conclusions about
potential environmental impact of urban growth. But as discussed earlier its
availability can be highly variable.

62

References:
Alberti, M. and Marzluff, J.M., 2004. Ecological resilience in urban
ecosystems: linking urban patterns to human and ecological functions.
Urban ecosystems 7(3): 241-265.
Alcaraz-Segura, D., Di Bella, C.M., and Straschnoy, J.V. (eds), 2013. Earth
Observation of Ecosystem Services. CRC Press.
Aplin, P., 2005. Remote sensing: ecology. Progress in Physical Geography
29(1): 104-113.
Aspinall, R. and Pearson, D., 2000. Integrated geographical assessment of
environmental condition in water catchments: Linking landscape ecology,
environmental modelling and GIS. Journal of Environmental Management
59: 299-319.
Ayanu, Y.Z., Conrad, C., Nauss, T., Wegmann, M., and Koellner, T., 2012.
Quantifying and mapping ecosystem services supplies and demands: a
review of remote sensing applications. Environmental science & technology
46(16): 8529-8541.
Ban, Y., Hu, H. and Rangel, I. 2010. Fusion of QuickBird MS and
RADARSAT-1 SAR Data for Land-Cover Mapping: Object-based and
Knowledge-Based Approach. International Journal of Remote Sensing
31(6): 1391-1410.
Ban, Y., and Jacob, A., 2013. Object-based fusion of multitemporal
multiangle ENVISAT ASAR and HJ-1B multispectral data for urban landcover mapping. Geoscience and Remote Sensing, IEEE Transactions on,
51(4): 1998-2006.
Ban, Y., Jacob, A., and Gamba, P., 2014(a). Spaceborne SAR data for global
urban mapping at 30m resolution using a robust urban extractor. ISPRS
Journal of Photogrammetry and Remote Sensing (in press).
Ban, Y., and Wu, Q., 2005. RADARSAT SAR data for landuse/land-cover
classification in the rural-urban fringe of the Greater Toronto Area. In
Proceedings of 8th AGILE Conference on Geographic Information Science:
26-28.
Ban, Y., Yousif, O. and Hu, H., 2014(b). Fusion of SAR and Optical Data
for Urban Land Cover Mapping and Change Detection. In Global Urban

63

Monitoring and Assessment through Earth Observation. Weng, Q. (ed).

CRC Press.
Baraldi, A., and Parmiggiani, F., 1995. An investigation of the textural
characteristics associated with gray level cooccurrence matrix statistical
parameters. Geoscience and Remote Sensing, IEEE Transactions on, 33(2):
293-304.
Barnsley, M.J., and Barr, S.L., 2000. Monitoring urban land use by earth
observation. Surveys in Geophysics 21(2): 269289.
Barthel, S. and Elmqvist, J.N., 2007. The history of allotment areas in 20th
century Sweden and their role in maintaining ecosystem services in urban
areas. Presented at Seventh International Conference on Urban History:
European City in Comparative Perspective, Panteion University, Athens Piraeus, Greece, 27-30 October 2004, Session: Green spaces in cities since
1918: politics, ideologies, and perceptions.
Behling, R., Bochow, M., Foerster, S., Roessner, S., and Kaufmann, H.,
2015. Automated GIS-based derivation of urban ecological indicators using
hyperspectral remote sensing and height information. Ecological Indicators
48: 218-234.
Bino, G., Levin, N., Darawshi, S., Van der Hal, N., Reich-Solomon, A., and
Kark, S., 2008. Accurate prediction of bird species richness patterns in an
urban environment using Landsat-derived NDVI and spectral unmixing.
International Journal of Remote Sensing 29: 36753700.
Bissonette, J.A. (ed), 1997. Wildlife and landscape ecology: effects of
pattern and scale. Springer.
Blaschke, T., 2010. Object based image analysis for remote sensing. ISPRS
journal of photogrammetry and remote sensing 65(1): 2-16.
Boggs, G.S., 2010. Assessment of SPOT 5 and QuickBird remotely sensed
imagery for mapping tree cover in savannas. International Journal of
Applied Earth Observation and Geoinformation 12(4): 217-224.
Bolund, P., and Hunhammar, S., 1999. Ecosystem services in urban areas.
Ecological Economics 29(2): 293-301.
Botequilha Leito, A. and Ahern, J., 2002. Applying landscape ecological
concepts and metrics in sustainable landscape planning. Landscape and
Urban Planning 59: 65-93.

64

Burkhard, B., Kroll, F., Mller, F., and Windhorst, W., 2009. Landscapes
capacities to provide ecosystem servicesa concept for land-cover based
assessments. Landscape online 15(1): 22.
Butusov, O.B., 2003. Textural classification of forest types from Landsat 7
imagery, Mapping Sciences and Remote Sensing 40: 91104.
Chee, Y.E., 2004. An ecological perspective on the valuation of ecosystem
services. Biological conservation 120(4): 549-565.
Chen, M., Su, W., Li, L., Zhang, C., Yue, A., and Li, H., 2009. Comparison
of Pixel-based and Object-oriented knowledge-based Classification Methods
using SPOT5 Imagery. WSEAS Transactions on Information Science and
Applications 3(6): 477-489.
Cihlar, J., 2000. Land cover mapping of large areas from satellites: status
and research priorities. International Journal of Remote Sensing 21: 10931114.
Clausi, D.A., 2002. An analysis of co-occurrence texture statistics as a
function of grey level quantization. Canadian Journal of remote sensing
28(1): 45-62.
Cleve, C., Kelly, M., Kearns, F.R., and Moritz, M., 2008. Classification of
the wildlandurban interface: A comparison of pixel- and object-based
classifications using high-resolution aerial photography. Computers,
Environment and Urban Systems 32: 317326.
Conchedda, G., Durieux, L., and Mayaux, P., 2008. An object-based method
for mapping and change analysis in mangrove ecosystems. ISPRS Journal of
Photogrammetry and Remote Sensing 63(5): 578-589.
Cortes, C. and Vapnik, V.N., 1995. Support-Vector Networks. Machine
Learning 20(3): 273-297.
Costanza, R., dArge, R., de Groot, R., Farber, S., Grasso, M., Hannon, B.,
Limburg, K. Naeem, S., ONeill, R. V., Paruelo, J., Raskin, R. G., Sutton, P.
and van den Belt, M., 1997. The value of the worlds ecosystem services and
natural capital. Nature 387(15): 253260.
Costanza, R., and Folke, C., 1997. Valuing ecosystem services with
efficiency, fairness, and sustainability as goals. In Daily, G. (ed), Natures

65

services: societal dependence on natural ecosystems. Island Press,

Washington, D.C.: 49-70.
Daily, G.C., 1997. Natures Services: Societal Dependence on Natural
Ecosystems. Island Press, Washington, D.C., USA.
Darwish, A., Leukert, K., and Reinhardt, W., 2003. Urban land-cover
classification: an object based perspective. In 2nd GRSS/ISPRS Joint
Workshop on Remote Sensing and Data Fusion over Urban Areas, 278-282.
IEEE.
De Groot, R.S., 1992. Functions of Nature: Evaluation of Nature in
Environmental Planning, Management and Decision Making. WoltersNoordhoff BV, Groningen, The Netherlands.
De Groot, R.S., Alkemade, R., Braat, L., Hein, L., and Willemen, L., 2010.
Challenges in integrating the concept of ecosystem services and values in
landscape planning, management and decision making. Ecological
Complexity 7(3): 260-272.
De Groot, R.S., Wilson, M.A. and Boumans, R.M.J., 2002. A typology for
classification, description and valuation of ecosystem functions, goods and
services. Ecological Economics 41(3): 393408.
De Martino, M., Causa, F., and Serpico, S.B., 2003. Classification of optical
high resolution images in urban environment using spectral and textural
information. In Geoscience and Remote Sensing Symposium, 2003.
IGARSS'03. Proceedings. 2003 IEEE International 1: 467469.
de Sherbinin, A., Levy, M.A., Zell, E., Weber, S., and Jaiteh, M., 2014.
Using satellite data to develop environmental indicators. Environmental
Research Letters 9(8): 084013.
Dimitrakopoulos, K., Gitas, I.Z., Polychronaki, A., Katagis, T., and
Minakou, C., 2010. Land Cover/Use Mapping Using Object Based
Classification of SPOT Imagery. Remote Sensing for Science, Education,
and Natural and Cultural Heritage, Rainer Reuter (ed), EARSeL 2010.
Dixon, B., and Candade, N., 2008. Multispectral landuse classification using
neural networks and support vector machines: one or the other, or both?
International Journal of Remote Sensing 29(4): 1185-1206.
Dunn, C.P., Sharpe, D.M., Guntenspergen, G.R., Stearns, F., and Yang, Z.,
1991. Methods for analyzing temporal changes in landscape pattern. In:

66

Turner, M.G., Gardner, R.H. (eds), Quantitative methods in landscape

ecology: the analysis and interpretation of landscape heterogeneity,
Springer-Verlag, New York, pp 173-198.
Duro, D.C., Franklin, S.E., and Dub, M.G., 2012. A comparison of pixelbased and object-based image analysis with selected machine learning
algorithms for the classification of agricultural landscapes using SPOT-5
HRG imagery. Remote Sensing of Environment 118: 259-272.
Feng, X., Fu, B., Yang, X., and L, Y., 2010. Remote sensing of ecosystem
services: An opportunity for spatially explicit assessment. Chinese
Geographical Science 20(6): 522-535.
Forman, R., and Godron, M., 1986. Landscape Ecology. Wiley, New York.
Frank, S., Frst, C., Koschke, L., and Makeschin, F., 2012. A contribution
towards a transfer of the ecosystem service concept to landscape planning
using landscape metrics. Ecological Indicators 21: 30-38.
Franklin, S.E., and Peddle, D.R., 1990. Classification of SPOT HRV
imagery and texture features, International Journal of Remote Sensing, 11:
551556.
Franklin, S.E., and Wulder, M.A., 2002. Remote sensing methods in medium
spatial resolution satellite data land cover classification of large areas.
Progress in Physical Geography 26(2): 173-205.
Gamba, P. and Aldrighi, M., 2012. SAR Data Classification of Urban Areas
by Means of Segmentation Techniques and Ancillary Optical Data. IEEE
Journal of Selected Topics in Applied Earth Observations and Remote
Sensing 5(4): 11401148.
Gao, Y., Wu, Z., Lou, Q., Huang, H., Cheng, J., and Chen, Z., 2012.
Landscape ecological security assessment based on projection pursuit in
Pearl River Delta. Environmental monitoring and assessment 184(4): 23072319.
Gerard, F., Petit, S., Smith, G., Thomson, A., Brown, N., Manchester, S.,
Wadsworth, R., Bugar, G., Halada, L., Bezak, P., Boltiziar, M., De badts, E.,
Halabuk, A., Mojse, M., Petrovic, F., Gregor, M., Hazeu, G., Mucher, C.A.,
Wachowicz, M., Huitu, H., Tuominen, S., Kohler, R., Olschofsky, K.; Ziese,
H., Kolar, J., Sustera, J., Luque, S., Pino, J., Pons, X., Roda, F., Roscher, M.,
and Feranec, J. (2005). Linking pan-European land cover change to
pressures on biodiversity-Biopress final report 1st January 2003-31st

67

December 2005, sections 5 and 6. Progress in Physical Geography 34(2):

183-205.
Gibson, C.C., Ostrom, E., and Ahn, T.K., 2000. The concept of scale and the
human dimensions of global change: a survey. Ecological economics 32(2):
217-239.
Gong, P., Marceau, D.J., and Howarth, P.J., 1992. A comparison of spatial
feature extraction algorithms for land-use classification with SPOT HRV
data, Remote Sensing of Environment 40: 137151.
Gneralp, B. and Seto, K. C., 2013. Futures of global urban expansion:
uncertainties and implications for biodiversity conservation. Environmental
Research Letters 8(1): 014025.
Haas, J., and Ban, Y., 2014. Urban growth and environmental impacts in
Jing-Jin-Ji, the Yangtze, River Delta and the Pearl River Delta. International
Journal of Applied Earth Observation and Geoinformation 30: 42-55.
Haase, D., Larondelle, N., Andersson, E., Artmann, M., Borgstrm, S.,
Breuste, J., Gomez-Baggethun, E., Gren, ., Hamstead, Z., Hansen, R.,
Kabisch, N., Kremer, P., Langemeyer, J., Rall, E.L., McPhearson, T.,
Pauleit, S., Qureshi, S., Schwarz, N., Voigt, A., Wurster, D. and Elmqvist,
T., 2014. A quantitative review of urban ecosystem service assessments:
Concepts, models, and implementation. Ambio 43(4): 413-433.
Haines-Young, R., Green, D.R., and Cousins, S.H. 1993. Landscape ecology
and GIS. Taylor and Francis, London, 288 pp.
Haines-Young, R. and Potschin, M., 2013. Common International
Classification of Ecosystem Services (CICES): Consultation on Version 4,
August-December 2012. EEA Framework Contract No EEA/IEA/09/003
Halada, L., Jongman, R.G.H., Gerard, F., Whittaker, L., Bunce, R.G.H.,
Bauch, B., and Schmeller, D.S., 2009. The European Biodiversity
Observation Network-EBONE. In: Hebek, J., Hradec, J., Pelikn, E.,
Mrovsk, O., Pillmann, W., Holoubek, I. and Bandholtz, T. (eds),
Proceedings of European conference of the Czech Presidency of the Council
of the EU TOWARDS eENVIRONMENT. Challenges of SEIS and SISE:
Integrating Environmental Knowledge in Europe: 177-188. Masaryk
University.
Hansen, M.J., Franklin, S.E., Woudsma, C.G., and Peterson, M., 2001.
Caribou habitat mapping and fragmentation analysis using Landsat MSS,

68

TM, and GIS data in North Columbia Mountains, British Columbia, Canada.
Remote Sensing of Environment 77(1): 50-65.
Haralick, R.M., Shanmugam, K., and Dinstein, I.H., 1973. Textural features
for image classification. IEEE Transactions on systems, man and cybernetics
3(6): 610621.
Hargis, C.D., Bissonette, J.A., and David, J.L., 1998. The behavior of
landscape metrics commonly used in the study of habitat fragmentation.
Landscape Ecology 13(3): 167186.
Herold, M., 2009. Some Recommendations for Global Efforts in Urban
Monitoring and Assessments from Remote Sensing. In Global Mapping of
Human Settlement: Experiences, Datasets, and Prospects. Gamba, P., and
Herold, M. (eds). Taylor and Francis Press.
Herold, M., Couclelis, H., and Clarke, K.C., 2005. The role of spatial metrics
in the analysis and modeling of urban land use change. Computers,
Environment and Urban Systems 29(4): 369399.
Herold, M., Liu, X., and Clarke, K.C., 2003. Spatial metrics and image
texture for mapping urban land use. Photogrammetric Engineering and
Remote Sensing 69(9): 9911002.
Herold, M., Schiefer, S., Hostert, P. and Roberts D.A., 2007. Applying
Imaging Spectrometry in Urban Areas. In Urban Remote Sensing. Weng, Q.
and Quattrochi, D.A. (eds). CRC Press, Taylor & Francis Group.
Hu, H., and Ban, Y., 2012. Multitemporal RADARSAT-2 ultra-fine beam
SAR data for urban land cover classification. Canadian Journal of Remote
Sensing 38(1): 1-11.
Hu, Y., Ban, Y., Zhang, Q. and Liu, J., 2009. The Trajectory of Urbanization
Process in the Yangtze River Delta During 1990 to 2005. 2009 Joint Urban
Remote Sensing Event 18.
Huang, C., Davis, L.S. and Townshend, J.R.G., 2002. An assessment of
support vector machines for land cover classification. International Journal
of Remote Sensing 23(4): 725-749.
Huang, Y.C., Li, P., Zhang, L., and Zhong, Y., 2009. Urban land cover
mapping by spatial-spectral feature analysis of high resolution hyperspectral
data with decision directed acyclic graph SVM. In Urban Remote Sensing
Event, 2009 Joint IEEE 1-7.

69

Imhoff, M.L., Zhang, P., Wolfe, R.E., and Bounoua, L., 2010. Remote
sensing of the urban heat island effect across biomes in the continental USA.
Remote Sensing of Environment 114(3): 504-513.
Jacobs, S., Burkhard, B., Van Daele, T., Staes, J., and Schneiders, A., 2014.
The Matrix Reloaded: A review of expert knowledge use for mapping
ecosystem services. Ecological Modelling 295: 21-30.
Jacquin, A., Misakova, L. and Gay, M., 2008. A hybrid object-based
classification approach for mapping urban sprawl in periurban environment.
Landscape and Urban Planning 84: 152165.
Jebur, M.N., Mohd Shafri, H.Z., Pradhan, B., and Tehrany, M.S., 2013. Perpixel and object-oriented classification methods for mapping urban land
cover extraction using SPOT 5 imagery. Geocarto International 29(7): 792806.
Jensen, J.R., 2005. Introductory Digital Image Processing, 3rd Ed., Upper
Saddle River, NJ: Prentice Hall, 526 pp.
Jin, S., Li, D. and Gong, J., 2005. A comparison of SVMs with MLC
algorithms on texture features. In MIPPR 2005 Image Analysis Techniques
Proc. of SPIE Vol. 6044 60442B-1.
Kamusoko, C., and Aniya, M., 2007. Land Use/Cover Change and
Landscape Fragmentation Analysis in the Bindura District, Zimbabwe. Land
Degradation & Development 18: 221-233.
Kasanko, M., Sagris, V., Lavalle, C., 2007(a). Analysing the compactness of
urban areas by using indicators derived from data acquired by remote
sensing. IEEE: 2007 Urban Remote Sensing Joint Event.
http://ieeexplore.ieee.org/xpls/abs_all.jsp?arnumber=4234438. Accessed
2014-11-19
Kasanko, M., Sagris, V., Lavalle, C., Barredo, J.I., Petrov, L., Steinnocher,
K., Loibl, W., Hoffmann, C., 2007(b). GEOLAND spatial planning
observatory: How remote sensing data can serve the needs of urban and
regional planning. IEEE: 2007 Urban Remote Sensing Joint Event.
http://ieeexplore.ieee.org/xpls/abs_all.jsp?arnumber=4234404. Accessed
2014-11-19

70

Kershaw, C.D., and Fuller, R.M., 1992. Statistical problems in the

discrimination of land cover from satellite images: a case study in lowland
Britain. International Journal of Remote Sensing 13: 3085-3104.
Klemas, V.V., 2001. Remote Sensing of Landscape-Level Coastal
Environmental Indicators. Environmental Management 27(1): 47-57.
Konarska, K.M., Sutton, P.C., and Castellon, M., 2002. Evaluating scale
dependence of ecosystem service valuation: a comparison of NOAAAVHRR and Landsat TM datasets. Ecological Economics 41(3): 491507.
Krishnaswamy, J., Bawa, K.S., Ganeshaiah, K.N., and Kiran, M.C., 2009.
Quantifying and mapping biodiversity and ecosystem services: Utility of a
multi-season NDVI based Mahalanobis distance surrogate. Remote Sensing
of Environment 113(4): 857-867.
Lambin, E.F., Turner, B.L., Geist, H.J., Agbola, S.B., Angelsen, A., Bruce,
J.W., Coomes, O.T., Dirzo, R., Fischer, G., Folke, C., George, P.S.,
Homewood, K., Imbernon, J., Leemans, R., Li, X., Moran, E.F., Mortimore,
M., Ramakrishnan, P.S., Richards, J.F., Sknes, H., Steffen, W., Stone, G.D.,
Svedin, U., Veldkamp, T.A., Vogel, C., and Xu, J., 2001. The causes of
land-use and land-cover change: moving beyond the myths. Global
Environmental Change 11: 261269.
Lakes, T., and Kim, H.O., 2012. The urban environmental indicator
Biotope Area RatioAn enhanced approach to assess and manage the
urban ecosystem services using high resolution remote-sensing. Ecological
Indicators 13(1): 93-103.
Leito, A.B., Miller, J., Ahern, J., and McGarigal, K., 2006. Measuring
landscapes: A planner's handbook. Island press.
Li, C., Wang, J., Wang, L., Hu, L., and Gong, P., 2014. Comparison of
Classification Algorithms and Training Sample Sizes in Urban Land
Classification with Landsat Thematic Mapper Imagery. Remote Sensing
6(2): 964-983.
Li, H. and Reynolds, J.F., 1994. A simulation experiment to quantify spatial
heterogeneity in categorical maps. Ecology 75: 2446-2455.
Li, H. and Wu, J., 2004. Use and misuse of landscape indices. Landscape
Ecology 19: 389-399.

71

Li, S., Sui, Y., Sun, Z., Wang, F., and Li, Y., 2005. Landscape pattern and
fragmentation of natural secondary forests in the eastern mountainous
region, northeast China: A case study of Mao'ershan forests in Heilongjiang
Province. Journal of Forestry Research 16(1): 35-38.
Li, Y., Zhu, X., Sun, X. and Wang, F., 2010. Landscape effects of
environmental impact on bay-area wetlands under rapid urban expansion and
development policy: a case study of Lianyungang, China. Landscape and
urban planning 94(3): 218-227.
Liang, B. and Weng, Q., 2011. Assessing urban environmental quality
change of Indianapolis, United States, by the remote sensing and GIS
integration. Selected Topics in Applied Earth Observations and Remote
Sensing, IEEE Journal of, 4(1): 43-55.
Lindenmayer, D.B., and Fischer, J., 2006. Habitat Fragmentation and
Landscape Change. Island Press, Washington.
Lisita, A., Sano, E.E., and Durieux, L., 2013. Identifying potential areas of
Cannabis sativa plantations using object-based image analysis of SPOT-5
satellite data. International Journal of Remote Sensing 34(15): 5409-5428.
Liu, X.H., Skidmore, A.K., and Van Oosten, H., 2002. Integration of
classification methods for improvement of land-cover map accuracy. ISPRS
Journal of Photogrammetry and Remote Sensing 56(4): 257268.
Lo, C.P., 1998. Application of Landsat TM Data for Quality of Life
Assessment in an Urban Environment. Computers, Environment and Urban
Systems 21(3/4): 259-276.
Long, J.A., Nelson, T.A., and Wulder, M.A., 2010. Characterizing forest
fragmentation: Distinguishing change in composition from configuration.
Applied Geography 30(3): 426-435.
Lfvenhaft, K., Runborg, S. and Sjgren-Gulve, P., 2004. Biotope patterns
and amphibian distribution as assessment tools in urban landscape planning.
Landscape and urban planning 68(4): 403-427.
Lu, D., and Weng, Q., 2005. Urban classification using full spectral
information of Landsat ETM+ imagery in Marion County, Indiana.
Photogrammetric engineering and remote sensing 71(11): 12751284.
Ludwig, D., 2000. Limitations of economic valuation of ecosystems.
Ecosystems 3(1): 31-35.

72

MA (Millennium Ecosystem Assessment), 2005. Ecosystems and Human

Well-being: Synthesis. Washington DC: Island Press.
Mathur, A., and Foody, G.M., 2008. Multiclass and binary SVM
classification: Implications for training and classification users. Geoscience
and Remote Sensing Letters, IEEE 5(2): 241-245.
McGarigal, K., 2002. Landscape pattern metrics. In El-Shaarawi, A.H.,
Piegorsch, W.W. (eds), Encyclopedia of environmetrics, 2. John Wiley &
Sons, Chichester, England, pp. 1135-1142.
McGarigal, K., and Marks, B.J., 1995. FRAGSTATS: Spatial pattern
analysis program for quantifying landscape structure. Gen. Tech. Rep. PNWGTR-351. US Department of Agriculture, Forest Service, Pacific Northwest
Research Station.
McGarigal, K. and McComb, W.C., 1995. Relationships between landscape
structure and breeding birds in the Oregon Coast Range. Ecological
monographs 65(3): 235-260.
Metria GeoAnalys, 2009. Ny metod fr kontinuerlig naturtypskartering av
skyddade omrden (KNAS). Rapport fr Naturvrdsverket.
Michaud, J.S., Coops, N.C., Andrew, M.E., Wulder, M.A., Brown, G.S., and
Rickbeil, G.J., 2014. Estimating moose (Alces alces) occurrence and
abundance from remotely derived environmental indicators. Remote Sensing
of Environment 152: 190-201.
Mountrakis, G., Im, J., and Ogole, C., 2011. Support vector machines in
remote sensing: A review. ISPRS Journal of Photogrammetry and Remote
Sensing 66(3): 247-259.
Mrtberg, U.M., Balfors, B. and Knol, W.C., 2007. Landscape ecological
assessment: a tool for integrating biodiversity issues in strategic
environmental assessment and planning. Journal of Environmental
Management 82(4): 457.
Narumalani, S., D.R. Mishra and R.G. Rothwell, 2004. Change detection and
landscape metrics for inferring anthropogenic processes in the greater EFMO
area. Remote Sensing of Environment 91(3-4): 478-489.
Nelson, E., Mendoza, G., Regetz, J., Polasky, S., Tallis, H., Cameron, D.R.,
Chan, K.M., Daily, G.C., Goldstein, J., Kareiva, P.M., Lonsdorf, E., Naidoo,

73

R., Ricketts, T.H., and Shaw, M.R., 2009. Modeling multiple ecosystem
services, biodiversity conservation, commodity production, and tradeoffs at
landscape scales. Frontiers in Ecology and the Environment 7(1): 4-11.
Newman, M.E., McLaren, K.P., and Wilson, B.S., 2011. Comparing the
effects of classification techniques on landscape-level assessments: pixelbased versus object-based classification. International Journal of Remote
Sensing 32(14): 4055-4073.
Newton, A. C., Hill, R. A., Echeverra, C., Golicher, D., Benayas, J. M. R.,
Cayuela, L., and Hinsley, S. A., 2009. Remote sensing and the future of
landscape ecology. Progress in Physical Geography 33(4): 528-546.
Nichol, J.E., and Wong, M.S., 2007. Assessing Urban Environmental
Quality with Multiple Parameters. In Urban Remote Sensing. Weng, Q., and
Quattrochi, D.A. (eds), CRC Press, Taylor & Francis Group.
Niemeijer, D., and de Groot, R.S., 2008. A conceptual framework for
selecting environmental indicator sets. Ecological indicators 8(1): 14-25.
Niu, X., and Ban, Y., 2013. Multi-temporal RADARSAT-2 polarimetric
SAR data for urban land-cover classification using an object-based support
vector machine and a rule-based approach. International Journal of Remote
Sensing 34(1): 1-26.
NRC, 2007. Geoscape Toronto: Oak Ridges Moraine. Natural Resources
Canada (NRC), Government of Canada.
http://geoscape.nrcan.gc.ca/toronto/moraine_e.php (accessed on 2011-0421).
Olff, H. and Ritchie, M.E., 2002. Fragmented nature: consequences for
biodiversity. Landscape and urban planning 58(2): 83-92.
O'Neill, R.V., Krummel, J.R., Gardner, R.H., Sugihara, G., Jackson, B.,
DeAngelis, D.L., Milne, B.T., et al., 1988. Indices of landscape pattern.
Landscape ecology 1(3): 153162.
Petit, S., Firbank, L., Wyatt, B., & Howard, D. (2001). MIRABEL: models
for integrated review and assessment of biodiversity in European landscapes.
AMBIO: A Journal of the Human Environment 30(2): 81-88.
Qin, Y., Niu, Z., Chen, F., Li, B., and Ban, Y., 2013. Object-based land
cover change detection for cross-sensor images. International Journal of
Remote Sensing 34(19): 6723-6737.

74

Raciti, S.M., Hutyra, L.R., and Newell, J.D., 2014. Mapping carbon storage
in urban trees with multi-source remote sensing data: Relationships between
biomass, land use, and demographics in Boston neighborhoods. Science of
The Total Environment 500: 72-83.
Radoux, J., and Defourny, P., 2007. A quantitative assessment of boundaries
in automated forest stand delineation using very high resolution imagery.
Remote Sensing of Environment 110(4): 468-475.
Regionplane- och Trafikkontoret (RTK), 2008. Grnstruktur och landskap i
regional utvecklingsplanering. Rapport 9:2008. Accessed on 2014-11-19
from
http://www.tmr.sll.se/publikationer/2008/2008-9-Gronstruktur-ochlandskap-i-regional-utvecklingsplanering/
Regionplane- och Trafikkontoret (RTK), 2002. Regional Utvecklingsplan
2001 fr Stockholmsregionen. Program & frslag 2. ISSN 1402-1331. ISBN
91-86-57462-0. Accessed on 2014-11-18 from http://www.tmr.sll.se/MOSSdokument/Publikation/Publikationer_PoF_2002_2_RUFS_2001_sid_112.pdf
Reis, J.P., Silva, E.A., and Pinho, P., 2014. A review of spatial metrics for
urban growth and shrinkage. In De Silva, E., Healey, P., Harris, N., and Van
den Broeck, P. (eds). Research Methods in Spatial and Regional Planning,
Routledge: 279-292.
Ren, W., Zhong, Y., Meligrana, J., Anderson, B., Watt, W.E., Chen, J., and
Leung, H.L., 2003. Urbanization, land use, and water quality in Shanghai:
19471996. Environment International 29(5): 649-659.
Revenga, C., 2005. Developing indicators of ecosystem condition using
geographic information systems and remote sensing. Regional
Environmental Change 5(4): 205-214.
Riitters, K.H., O'Neill, R.V., Hunsaker, C.T., Wickham, J.D., Yankee, D.H.,
Timmins, S.P., Jones, K.B., and Jackson, B.L., 1995. A factor analysis of
landscape pattern and structure metrics. Landscape Ecology 10(1): 2339.
Rodriguez-Galiano, V.F., Chica-Olmo, M., Abarca-Hernandez, F., Atkinson,
P. M., and Jeganathan, C., 2012. Random Forest classification of
Mediterranean land cover using multi-seasonal imagery and multi-seasonal
texture. Remote Sensing of Environment 121: 93-107.

75

Rozenstein, O., and Karnieli, A., 2011. Comparison of methods for land-use
classification incorporating remote sensing and GIS inputs. Applied
Geography 31(2): 533-544.
Saaty, T.L., 2001. Decision Making for Leaders: The Analytic Hierarchy
Process for Decisions in a Complex World. 3rd ed., RWS Publications,
ISBN 0-9620317-8-X.
Seto, K. C., Gneralp, B., and Hutyra, L. R., 2012. Global forecasts of urban
expansion to 2030 and direct impacts on biodiversity and carbon pools.
Proceedings of the National Academy of Sciences 109(40): 16083-16088.
Seto, K.C., Sanchez-Rodriguez, R., and Fragkias, M., 2010. The New
Geography of Contemporary Urbanization and the Environment. Annual
Review of Environment and Resources 35: 167194.
Shaban, M.A., and Dikshit, O., 2001. Improvement of classification in urban
areas by the use of textural features: the case study of Lucknow city, Uttar
Pradesh, International Journal of Remote Sensing 22: 565593.
Smeets, E. and Weterings, R., 1999. Environmental Indicators: Typology
and Overview. EEA Technical Report no. 25. Copenhagen: European
Environment Agency.
Srivastava, P.K., Han, D., Ramirez, M.R., Bray, M., and Islam, T., 2012.
Selection of classification techniques for land use/land cover change
investigation. Advances in Space Research 50: 1250-1265.
Stefanov, W.L., Ramsey, M.S., and Christensen, P.R., 2001. Monitoring
urban land cover change: An expert system approach to land cover
classification of semiarid to arid urban centers. Remote Sensing of
Environment 77: 173-185.
Sthl, G., Allard, A., Esseen, P.-A., Glimskr, A., Ringvall, A., Svensson, J.,
Sundquist, S., Christensen, P., Gallegos Torell, ., Hgstrm, M.,
Lagerqvist, K., Marklund, L., Nilsson, B. and Inghe, O., 2011. National
Inventory of Landscapes in Sweden (NILS) scope, design, and
experiences from establishing a multiscale biodiversity monitoring system.
Environmental Monitoring and Assessment 173(14): 579595.
Statistics Sweden, 2008. Grnytor i ttort. Satellitdata som std vid kartering
av grnytor i och omkring ttorter. DNR (Rymdstyrelsen): 180/06.

76

Stockholms Stads Utrednings- och Statistikkontor AB (USK) 2011. Statistisk

rsbok fr Stockholm 2011. Accessed on 2014-11-18 from
http://www.statistikomstockholm.se/index.php/publikationer/statistiskarsbok-foer-stockholm
Su, S., Ma, X., and Xiao, R., 2014. Agricultural landscape pattern changes in
response to urbanization at ecoregional scale. Ecological Indicators 40: 1018.
Su, S., Xiao, R., Jiang, Z., and Zhang, Y., 2012. Characterizing landscape
pattern and ecosystem service value changes for urbanization impacts at an
eco-regional scale. Applied Geography 34: 295-305.
Su, W., Zhang, C., Zhu, X., and Li, D., 2009. A hierarchical object oriented
method for land cover classification of SPOT 5 imagery. WSEAS
Transactions on Information Science and Applications 6(3): 437-446.
Su, W., Zhang, C., Luo, M., Li, L., Wang, Y., Ju, Z., and Li, D., 2008.
Object oriented implementation monitoring method of zone feature in land
consolidation engineering using SPOT 5 imagery. WSEAS Transactions on
Computers 7(7): 847-856.
Tallis, H., Mooney, H., Andelman, S., Balvanera, P., Cramer, W., Karp, D.,
Polasky, S., Reyers, B., Ricketts, T., Running, S., Thonicke, K., Tietjen, B.
and Walz, A., 2012. A global system for monitoring ecosystem service
change. BioScience 62(11): 977-986.
Tallis, H., and Polasky, S., 2009. Mapping and valuing ecosystem services
as an approach for conservation and naturalresource management. Annals of
the New York Academy of Sciences, 1162(1): 265-283.
Tehrany, M.S., Pradhan, B., and Jebu, M.N., 2014. A comparative
assessment between object and pixel-based classification approaches for
land use/land cover mapping using SPOT 5 imagery. Geocarto International
29(4): 351-369.
Tiede, D., Moeller, M., Lang, S., and Hoelbling, D., 2007. Adapting,
splitting and merging cadastral boundaries according to homogenous LULC
types derived from SPOT 5 data. In Proc. of the ISPRS Workshop Photogr.
Image Analysis, Munich.
Toronto and Region Conservation, 2014. Regulation and Policy
Development. Accessed on 2014-11-19 from

77

http://www.trca.on.ca/planning-services-permits/regulation-and-policydevelopment.dot
Troy, A., and Wilson, M.A., 2006. Mapping ecosystem services: practical
challenges and opportunities in linking GIS and value transfer. Ecological
economics 60(2): 435-449.
Turner, M.G., 1989. Landscape Ecology: The Effect of Pattern on Process.
Annual Review of Ecology and Systematics 20: 171-197.
Turner, M.G., 1990. Spatial and temporal analysis of landscape patterns.
Landscape Ecology 4(1): 2130.
Turner, R.K., MorseJones, S., and Fisher, B., 2010. Ecosystem valuation.
Annals of the New York Academy of Sciences 1185(1): 79-101.
Turner, R.K., Paavola, J., Cooper, P., Farber, S., Jessamy, V., and Georgiou,
S., 2003. Valuing nature: lessons learned and future research directions.
Ecological economics 46(3): 493-510.
Uuemaa, E., Antrop, M., Roosaare, J., Marja, R., and Mander, ., 2009.
Landscape metrics and indices: an overview of their use in landscape
research. Living Reviews in Landscape Research 3(1): 1-28.
Uuemaa, E., Mander, ., and Marja, R., 2013. Trends in the use of landscape
spatial metrics as landscape indicators: a review. Ecological Indicators 28:
100-106.
UNEP/RIVM, 1994. An Overview of Environmental Indicators: State of the
art and perspectives. UNEP/EATR.94-01; RIVM/402001001. Environmental
Assessment Sub-Programme, UNEP, Nairobi.
United Nations, 2012. World Urbanization Prospects: The 2011 Revision,
United Nations Department of Economic and Social Affairs/Population
Division.
van Oudenhoven, A.P., Petz, K., Alkemade, R., Hein, L., & de Groot, R.S.,
2012. Framework for systematic indicator selection to assess effects of land
management on ecosystem services. Ecological Indicators 21: 110-122.
Vapnik, V.N., 1995. The Nature of Statistical Learning Theory. New York:
Springer-Verlag.

78

Wakelyn, L.A., 1990. Wetland inventory and mapping in the Northwest

Territories using digital Landsat data. Yellowknife: Government of the
Northwest Territories Department of Renewable Resources, File Report No.
96.
Wang, Z., Wei, W., Zhao, S., and Chen, X., 2004. Object-oriented
Classification and Application in Land Use Classification Using SPOT-5
PAN Imagery. 2004 IEEE International Geoscience and Remote Sensing
Symposium Proceedings 5: 3158-3160.
Weng, Q., 2002. Land use change analysis in the Zhujiang Delta of China
using satellite remote sensing, GIS and stochastic modelling. Journal of
Environmental Management 64: 273284.
Weng, Q, Gamba, P., Mountrakis, G., Pesaresi, M., Lu, L., Kemper, T.,
Heinzel, J., Xian, G., Jin, H., Miyazaki, H., Xu, B., Quresh, S.,
Keramitsoglou, I., Ban, Y., Esch, T., Roth, A. and Elvidge, C.D., 2014.
Urban Observing Sensors. In Global Urban Monitoring and Assessment
through Earth Observation. Weng, Q. (ed). CRC Press.
Wentz, E.A., Stefanov, W.L., Netzband, M., Mller, M.S. and Brazel, A.J.,
2009. The Urban Environmental Monitoring/100 Cities Project: Legacy of
the First Phase and Next Steps. In Global Mapping of Human Settlement:
Experiences, Datasets, and Prospects. Gamba, P., and Herold, M., (eds)
Taylor and Francis Press.
World Health Organization (WHO), 1999. Berglund, B., Lindvall, T., and
Schwela, D.H. (eds). Guidelines for Community Noise. Accessed on 201411-19 from http://www.who.int/docstore/peh/noise/Comnoise-1.pdf
Willemen, L., Verburg, P.H., Hein, L., and van Mensvoort, M.E., 2008.
Spatial characterization of landscape functions. Landscape and Urban
Planning 88(1): 34-43.
Wu, J., 2004. Effects of changing scale on landscape pattern analysis:
scaling relations. Landscape Ecology 19(2): 125-138.
Wu, J., Jenerette, G.D., Buyantuyev, A., and Redman, C.L., 2011.
Quantifying spatiotemporal patterns of urbanization: The case of the two
fastest growing metropolitan regions in the United States. Ecological
Complexity 8(1): 1-8.
Wu, K.Y., Ye, X.Y., Qi, Z.F., and Zhang, H., 2013. Impacts of land use/land
cover change and socioeconomic development on regional ecosystem

79

services: The case of fast-growing Hangzhou metropolitan area, China.

Cities 31: 276-284.
Yang, X. (ed), 2011. Urban Remote Sensing: Monitoring, Synthesis and
Modeling in the Urban Environment. ISBN: 978-0-470-74958-6. 408 pp.
Zhang, Q., Ban, Y., Liu, J., and Hu, Y., 2011. Simulation and analysis of
urban growth scenarios for the Greater Shanghai Area, China. Computers,
Environment and Urban Systems 35(2): 126-139.
Zhou, W., Troy, A., and Grove, M., 2008. Object-based land cover
classification and change analysis in the Baltimore metropolitan area using
multitemporal high resolution remote sensing data. Sensors 8(3): 1613-1636.

80

81

Appendix: Environmental Impact Indicator Results

Table 1. Environmental Impact Index Results in 1986 (Note that LFA 1 and 6 have greater areas than those given here. They extend
beyond the classification boundary and thus their full areas are not calculable. Due to this circumstance, they are given the same highest rank for area.)

ID
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15

ID parish/
municipality
Huddinge_
Botkyrka
Haninge
Boo
West sterker
West
Norrsunda
North
sterker
Trngsund
Skarpnck
East Nacka
Lov
Vaxholm
Sollentuna_
Jrflla
Danderyd_
Tby
Central
sterker
Ed_Kallhll

Area 2
km

Thickness
(m - radius)

500m width
(250m radius)?

Perforation
Indicator

Prox. to Intense
Urban land use
Indicator

% Urban
Perimeter

Noise (area
affected 2
km )

>189.4
128.4
11.5

883
1269
390

yes
yes
yes

20.45
15.53
7.97

615.6
425.6
19.5

0.069
0.20
0.185

20.34
1.55
1.18

17.0

976

yes

5.38

33.6

0.053

0.68

6.2

444

yes

3.76

24.1

0.070

3.92

>146.0
10.5

1159
467

yes
yes

15.40
5.29

703.8
22.0

0.074
0.138

4.28
0.65

3.2
14.7
4.4
20.8

371
793
420
615

yes
yes
yes
yes

4.76
5.97
4.46
9.43

5.3
24.0
6.3
31.1

0.12
0.1801
0.012
0.016

0.24
1.55
0.00
0.26

12.9

529

yes

6.92

19.6

0.008

2.81

9.0

382

yes

7.07

11.2

0.19

0.43

10.3
9.6

603
745

yes
yes

5.29
4.55

30.0
21.3

0.063
0.03

0.002
3.56

82

16
17
18
19
20
21
22
23
24
25
26
27
28
29
30

Ed
Norrsunda_
Vallentuna
Central
Vallentuna

6.5

515

yes

4.06

17.1

0.011

0.75

3.9

364

yes

4.76

13.0

0.00

2.55

10.6

497

yes

6.38

33.3

0.056

0.05

Sankt Mikael
East sterker
Flemingsberg
Huddinge
West Nacka
Sollentuna_
Fresta
North
Norrsunda
North
Vallentuna

10.2
12.2
4.0
3.7
7.0

720
607
357
445
534

yes
yes
yes
yes
yes

4.15
4.40
4.47
4.61
4.72

15.6
44.8
7.2
6.8
9.3

0.39
0.10
0.15
0.04
0.36

0.80
1.80
0.07
0.09
0.16

6.0

504

yes

4.44

9.4

0.11

1.06

18.0

598

yes

7.81

76.1

0.023

15.14

14.3

657

yes

6.97

64.6

0.00

0.53

Viksj
Jakobsberg_
Kallhll
Tby_
Vallentuna
East Vallentuna

4.4

447

yes

3.88

7.8

0.142

0.18

3.2

668

yes

2.24

5.8

0.1796

0.52

20.5
24.8

707
717

yes
yes

7.48
8.68

42.7
81.0

0.13
0.050

1.54
0.34

83

Table 2. Environmental Impact Ranking per Indicator and Average Rank for 1986
ID

ID parish/municipality
1
2
3
4
5

Prox. to Intense
Urban land use
Indicator

Perforation
Indicator

Area

Urban
Perimeter

Noise (area
affected)

Average Rank

Huddinge_Botkyrka
Haninge
Boo
West sterker
West Norrsunda

1
3
13
8
22

30
29
25
17
2

2
3
18
9
13

14
28
26
11
15

30
22
19
15
27

15.4
17.0
20.2
12.0
15.8

6
7
8
9
10

North sterker
Trngsund
Skarpnck
East Nacka
Lov

1
15
30
9
24

28
16
14
18
8

1
15
30
14
28

16
21
19
25
5

28
14
8
21
1

14.8
16.2
20.2
17.4
13.2

11
12
13
14

Vaxholm
Sollentuna_Jrflla
Danderyd_Tby
Central sterker

5
11
19
16

27
20
22
15

11
17
22
12

6
3
27
13

9
25
11
2

11.6
15.2
20.2
11.6

15
16
17
18
19

Ed_Kallhll
Ed
Norrsunda_Vallentuna
Central Vallentuna
Sankt Mikael

18
21
27
14
17

10
4
13
19
5

16
19
21
10
20

8
4
1
12
30

26
16
24
3
17

15.6
12.8
17.2
11.6
17.8

20
21

East sterker
Flemingsberg

12
26

6
9

7
26

17
23

23
4

13.0
17.6

84

22
23
24
25

Huddinge
West Nacka
Sollentuna_Fresta
North Norrsunda

28
20
23
7

11
12
7
24

27
24
23
5

9
29
18
7

5
6
18
29

16.0
18.2
17.8
14.4

26
27
28
29
30

North Vallentuna
Viksj
Jakobsberg_Kallhll
Tby_Vallentuna
East Vallentuna

10
25
29
6
4

21
3
1
23
26

6
25
29
8
4

1
22
24
20
10

13
7
12
20
10

10.2
16.4
19.0
15.4
10.8

85

Table 3. Environmental Impact Index Results in 2006 (Note that LFA 1.1 and 6 have greater areas than those given here. They extend
beyond the classification boundary and thus their full areas are not calculable. Due to this circumstance, they are given the same highest rank for area.)

Noise (area
affected 2
km )

>178.4

922

yes

22.41

3087

0.130

16.11

West Haninge
Haninge
Boo
West sterker

6.1
126.9
9.9
16.9

552
996
372
676

yes
yes
yes
yes

6.51
18.38
10.80
6.16

105
1081
76
131

0.303
0.298
0.475
0.248

0.31
1.62
0.87
0.61

5
6
7
8
9

West Norrsunda
North sterker
Trngsund
Skarpnck
East Nacka

7.6
>135.6
12.2
3.7
15.2

551
1127
1044
412
656

yes
yes
yes
yes
yes

3.89
20.10
5.28
5.06
7.27

79
1807
160
52
232

0.084
0.127
0.125
0.118
0.282

4.98
3.97
0.53
0.45
1.64

10
11

Lov
Vaxholm
Sollentuna_
Jrflla
Danderyd_Tby

5.7
24.7

624
760

yes
yes

5.60
8.22

96
350

0.056
0.058

0.00
0.49

13.1
10.5

542
417

yes
yes

6.81
7.67

156
113

0.052
0.232

2.97
0.53

6.7
7.1
3.3
7.4

593
602
364
609

yes
yes
yes
yes

4.69
4.15
4.804
4.468

61
69
30
56

0.410
0.207
0.032
0.018

0.00
3.63
0.30
0.67

3.6

371

yes

4.801

28

0.016

2.67

14
15.1
15.2
16
17

Central sterker
Ed_Kallhll
South Ed
Ed
Norrsunda_
Vallentuna

500m width
(250m radius)?

% Urban
Perimeter

1.2
2
3
4

12
13

Thickness
(m - radius)

Prox. to Intense
Urban land use
Indicator

1.1

ID

Area 2
km

Perforation
Indicator

ID parish/
municipality
Huddinde_
Botkyrka

86

18
19
20
21

Central Vallentuna
Sankt Mikael
East sterker
Flemingsberg

9.2
9.1
11.4
3.5

534
794
504
420

yes
yes
yes
yes

6.492
4.373
7.376
4.633

68
62
73
19

0.029
0.437
0.239
0.182

0.05
0.65
2.06
0.13

22
23
24
25
26

Huddinge
West Nacka
Sollentuna_Fresta
North Norrsunda
North Vallentuna

4.3
6.9
6.7
15.5
17.8

442
496
650
602
784

yes
yes
yes
yes
yes

4.630
6.37
3.96
7.54
8.13

36
58
93
148
188

0.038
0.306
0.075
0.068
0.001

0.10
0.22
1.21
12.67
0.22

27

3.5

447

yes

5.04

40

0.107

0.11

28
29
30.1

Viksj
Jakobsberg_
Kallhll
Tby_Vallentuna
East Vallentuna

3.1
10.2
12.4

607
535
467

yes
yes
yes

3.00
5.69
9.04

36
107
98

0.208
0.059
0.119

0.47
0.25
0.02

30.2

NE Vallentuna

7.7

462

yes

6.65

48

0.012

0.30

For comparability with 1986: (these values are used when comparing rankings between 1986 and 2006, instead of the true values for
LFA 1.1, 1.2, 15.1, 15.2, 30.1 and 30.2. These LFAs are penalized for becoming divided or fragmented by dropping one ranking spot in
the overall ranking, see below.)

ID
1
15
30

ID
parish/municipality
Huddinge_Botkyrka
Ed_Kallhll
East Vallentuna

(sum)
2
Area - km
184.5

(avr)
Perforation
Indicator
14.46

(avr) Prox. to
Intense Urban
land use Indicator
1595.8

10.4
20.1

4.48
7.85

49.5
73.0

87

0.22

(sum) Noise (area

affected by >55dB)
2
km
16.43

0.12
0.07

3.93
0.33

(avr) % Urban
Perimeter

Table 4. Environmental Impact Ranking per Indicator and Average Rank for 2006
ID
1.1
1.2
2
3
4
5
6
7
8
9
10
11
12
13
14
15.1
15.2
16
17
18
19
20

1
25

Perforation
Indicator
33
20

Haninge
Boo
West sterker
West Norrsunda
North sterker

3
15
6
19
1

Trngsund
Skarpnck
East Nacka
Lov
Vaxholm

Prox. to Intense
Urban land use
Indicator
1
13

20
29

Noise (area
affected)
33
13

31
30
17
2
32

3
18
10
17
2

28
33
26
14
19

24
22
19
31
30

17.80
23.60
15.60
16.60
16.80

11
28
8
26
4

14
13
23
15
28

7
26
5
15
4

18
16
27
9
10

17
14
25
1
16

13.40
19.40
17.60
13.20
12.40

Sollentuna_Jrflla
Danderyd_Tby
Central sterker
Ed_Kallhll
South Ed

9
13
23
21
32

22
26
9
4
11

8
11
23
20
31

8
24
31
22
6

28
18
2
29
11

15.00
18.40
17.60
19.20
18.20

Ed
Norrsunda_Vallentuna
Central Vallentuna
Sankt Mikael
East sterker

20
29
16
17
12

6
10
19
5
24

25
32
21
22
19

4
3
5
32
25

21
27
4
20
26

15.20
20.20
13.00
19.20
21.20

ID parish/municipality
Huddinde_Botkyrka
West Haninge

Area

88

Urban
Perimeter

Average Rank
17.60
20.00

21
22
23
24

Flemingsberg
Huddinge
West Nacka
Sollentuna_Fresta

30
27
22
24

8
7
18
3

33
29
24
16

21
7
30
13

7
5
9
23

19.80
15.00
20.60
15.80

25
26
27
28
29

North Norrsunda
North Vallentuna
Viksj
Jakobsberg_Kallhll
Tby_Vallentuna

7
5
31
33
14

25
27
12
1
16

9
6
28
30
12

12
1
15
23
11

32
8
6
15
10

17.00
9.40
18.40
20.40
12.60

East Vallentuna
NE Vallentuna

10
18

29
21

14
27

17
2

3
12

14.60
16.00

30.1
30.2

For comparability with 1986:

ID
1
15
30

ID parish/
municipality
Huddinge_
Botkyrka
Ed_Kallhll
East Vallentuna

Area
1
14
5

Perforation
28
6
24

Prox. to
Intense Urban
land use
2
25
18

89

Urban
Perimeter
21
16
10

Noise (area
affected)
30
26
10

Average
Rank
16.4
17.4
13.4

Adjusted Average
Rank (division
penalty)
17.4
18.4
14.4

Table 5. Change in indicator rankings and average ranking for Environmental Impact between 1986 and 2006
ID parish/
municipality

ID
1

Huddinge_Botkyrka

Haninge

Boo

4
5

Area 2
km
0

Perforation
Indicator

Prox. to Intense
Urban land use
Indicator

% Urban
Perimeter

Average
ranking
change

Noise (area
2
affected - km )

-7

Average ranking
change with
deduction for division
-1

-2

-3

-2

-4

-1

-1

West sterker

-1

-13

-1

-3

-3

West Norrsunda

-2

-1

North sterker

-2

-2

-1

-1

Trngsund

Skarpnck

-3

East Nacka

-2

-1

10

Lov

-6

15

-2

11

Vaxholm

-2

-4

12

Sollentuna_Jrflla

-3

13

Danderyd_Tby

-1

11

-4

14

Central sterker

-6

-9

-15

-5

-5

15

Ed_Kallhll

-9

-8

-2

-3

16

-1

-4

-2

-1

-1

17

Ed
Norrsunda
Vallentuna

-8

-1

-1

-1

18

Central Vallentuna

-3

-9

-1

-1

19

Sankt Mikael

-1

20

East sterker

-15

-10

-6

-6

-6

21

Flemingsberg

-2

-4

-2

-1

-1

22

Huddinge

90

23

West Nacka

-1

-5

24

Sollentuna_Fresta

25

North Norrsunda

-1

26

North Vallentuna

-4

27

-4

28

Viksj
Jakobsberg_
Kallhll

29
30

-2

-2

-4

-4

-1

-1

-8

-1

-1

-1

-1

Tby_Vallentuna

-9

-4

11

11

East Vallentuna

-1

-14

-3

-4

91

-1

-1

92