Journal of Planning Education and Research
© The Author(s) 2019
Article reuse guidelines:
In recent decades, the urban phenomenon has assumed an
increasingly important role in policies and strategies
designed to achieve sustainable development as cities have
become home to a sizeable and ever-growing proportion of
the world’s population (United Nations 2014). To address
the challenges of urban sustainability, the twenty-first-cen-
tury city needs to be seen as essentially metropolitan in
nature (Hall 1998; Organisation for Economic Co-operation
and Development [OECD] 2012). From a metropolitan per-
spective, the scale of social, economic, and environmental
processes transcends the boundaries of the traditional city,
for the most part surpassing the capacity for intervention of
individual municipalities. In terms of spatial planning,
urban, rural, and natural landscapes come together in a pro-
gressively widespread, hybrid, and ever-changing spatial
system which requires innovative research methods and
planning strategies capable of meeting the added complex-
ity derived from the upsizing of the urban phenomenon.
Metropolitan growth processes, especially, when associated
with urban sprawl, bring about major changes to landscape
structure and functionality. Some of the main consequences
include a loss of natural and agricultural areas, habitat frag-
mentation (Johnson and Klemens 2005; Marzluff and Ewing
2008), and a subsequent fall in ecosystem services (Reice
2005). Planning strategies should be capable not only of
responding to the current—essentially supramunicipal—
scale of these processes and impacts but also of addressing
the complex relationships established between the different
artificial and natural components of the metropolitan land-
scape. Both factors are crucial for achieving a sustainable
and well-adjusted balance between the city and its natural
and rural surroundings.
In this context, landscape spatial analysis methodologies
(Forman and Godron 1986) have shown to meet construct
validity criteria insofar as detecting and monitoring the main
trends of change in the metropolitan space and providing rel-
evant information for decision making and planning. These
methodologies have been widely applied to the study of
structural patterns in metropolitan areas, especially when it
comes to identifying land cover change dynamics (Aguilera,
Valenzuela, and André Botequilha-Leitão 2011; Shrestha
et al. 2012; Tian et al. 2011; Wu et al. 2011) and for the com-
parative analysis of different areas (Angel, Parent, and Civco
2012; Kaza 2013; Pijanowski and Robinson 2011; Yang
et al. 2012). In particular, methods based on landscape met-
rics are very useful for addressing aspects of nonbuilt space
and the effects of urbanization on spatial connectivity
845439JPEXXX10.1177/0739456X19845439Journal of Planning Education and ResearchFeria Toribio and Santiago Ramos
Initial submission, June 2016; revised submissions, May 2017, June and
October 2018, January 2019; final acceptance, March 2019
1Pablo de Olavide University, Seville, Spain
Jesús Santiago Ramos, Department of Geography, History and Philosophy,
Pablo de Olavide University, Ctra. de Utrera, km 1, 41013, Seville, Spain.
Landscape Spatial Analysis for
Sustainable Land Use Planning: A
Two-Scale Approach to the Seville
José María Feria Toribio1 and Jesús Santiago Ramos1
Metropolitan areas are complex, dynamic spatial systems. This paper sets out a spatial analysis methodology suitable to
address the internal complexity of metropolitan landscape change and which results could be useful for decision making in the
context of sustainable land use planning. A two-scale approach is adopted for the analysis of recent land use changes in the
metropolitan area of Seville (Spain), being the methodology applied to both the whole metropolitan area and two different
landscape units. Distinct landscape change patterns and urban growth models are identified for the units studied. On the basis
of the results, the convenience of a multiscale planning approach is highlighted.
applied GIS, land use planning, metropolitan areas, multiscale spatial analysis, urban sustainability
2 Journal of Planning Education and Research 00(0)
(Coskun 2013), landscape fragmentation (Dewan,
Yamaguchi, and Rahman 2012; Furberg and Ban 2012; La
Rosa and Privitera 2013), and the functionality of urban open
spaces as a habitat for animal species (Lizée et al. 2012;
Magle et al. 2012; Parrish and Hepinstall-Cymerman 2012).
They have also been applied to assess green infrastructure
planning initiatives (Lynch 2016).
In contrast, the application of study outcomes based on
landscape metrics to urban and regional planning practices
has yet to become widespread. One possible explanation lies
in the technical difficulties associated with selecting and
interpreting the metrics (Botequilha-Leitão and Ahern 2002);
however, it could also be related to the frequent disconnect
between the traditional practices employed by planners and
the theoretical and methodological contributions made in the
environmental sciences (Feria and Santiago 2009).
Nonetheless, it is necessary to step up efforts to develop
practice-oriented research methodologies that can be easily
adopted by planners. In the case of metropolitan planning,
these approaches must address the complex particularities of
these large urban systems and provide a synthetic, purpose-
ful, and comprehensive assessment of their spatial configura-
tion and evolution.
Starting from this premise, the paper is primarily aimed at
applying a landscape analysis methodology suitable for
addressing the internal complexity of metropolitan landscape
dynamics and whose results could prove useful for planning
purposes at a metropolitan scale. The basic assumption
underlying the analytical framework is that the spatial pat-
tern of urban development in a metro area is internally het-
erogeneous and dependent on several factors, including
landscape configuration and site-specific determinants. In
other words, patterns and processes of urban growth do not
generally follow a conventional center–periphery or Von
Thunen (urban–rural gradient) model, which can be
addressed by adopting a gradient or buffer analysis approach,
but they are strongly conditioned by the internal heterogene-
ity of the metropolitan landscape and urbanization models
(Forman 2008). Such heterogeneity has not been properly
addressed in metropolitan planning processes, at least in the
The hypothesis is tested using a two-scale approach, com-
paring the whole metropolitan area and two submetropolitan
landscape units. The central research question behind this
approach is as follows: to what extent do local landscape
change processes differ from general trends for the whole
metropolitan area and, therefore, require specific planning
guidelines? To this end, an analysis conducted on the Seville
Metropolitan Area is presented as a representative case.
Specifically, this analysis adopts an “embedded, single-case
design” (Yin 2002), as it meets the conditions for size, scale,
and structural complexity that enables a multiscale approach.
To test the approach’s internal validity, a comparative analy-
sis covering two different time periods (1984–1999 and
1999–2007) is performed, allowing us to ascertain the
intensity and impact of recent changes. The magnitude of
land use change processes in Spanish metropolitan areas
between 1984 and 2007 and their significant reduction fol-
lowing the 2008 economic crisis (Albertos and Sánchez
2014) justifies interest in this study period as it allows us to
establish causal relationships between dynamics of urban
change and landscape metrics. Two levels of thematic resolu-
tion are considered to analyze both general and specific
trends regarding the different components of the metropoli-
tan landscape. The analysis results will be discussed from a
spatial planning-oriented perspective to test the usefulness of
this methodological approach in supporting decision-making
processes and the design of strategies aimed at achieving
sustainable metropolitan development.
The external validity and reliability of the approach can be
tested with other metropolitan areas of a similar scale, urban
dynamism, a complex internal landscape configuration, and
different planning instruments. Well-known examples include
the Greater Copenhagen Metropolitan Area and its
Development Plan, widely known as the Finger Plan, which
represents a relevant application case study. In this region, the
main urban development processes were articulated around
the so-called city fingers—radial urban corridors linked to the
railway system and the radial road network—separated by a
system of green wedges preserved from urban development.
Another potential applied case study is the Stuttgart
Metropolitan Region, which encompasses a high urban den-
sity with a polycentric spatial configuration. Given its charac-
teristics, the planning area was divided into six differentiated
zones of regulation under the 2009 Regional Plan. A third and
final example is the Territorial Plan of the Barcelona
Metropolitan Region, an area that boasts high structural and
functional complexity in relation to both landscape configu-
ration and the display of settlement, economic, and infrastruc-
tural systems. All these areas share a common feature:
complex internal heterogeneity which, as in the case of
Seville, calls for a methodological alternative to the conven-
tional center–periphery analytic framework.
The study site, the Seville Metropolitan Area, covers a sur-
face area of approximately 4,900 km2 with 1,550,636 inhab-
itants according to the 2011 census. It represents the fourth
largest metropolitan area in Spain after Madrid, Barcelona,
and Valencia. The area’s delimitation, adopted by the
Metropolitan Plan approved in 2009, is based on conven-
tional and internationally standardized criteria of both func-
tional (commuting) and morphological (urban land use)
references (Feria and Martínez 2016; OECD 2012).
The metropolitan area is built around a central city, the
municipality of Seville. Only very recently has this histori-
cally large-in-size urban city seen urban growth processes
outside the limits of its municipal boundaries. In fact, the
metropolitan urban processes in the area are relatively recent
Feria Toribio and Santiago Ramos 3
compared with other Spanish and European cities and held
no major importance until the 1980s. However, from that
moment on, demographic growth and urban sprawl rates
have been especially intense. Over the last three decades,
population growth has always exceeded 10%, whereas the
area’s urbanized part as a whole has grown more than 50% in
size during said period (Feria 2018). The population distribu-
tion in the area is progressively bowing toward a very
dynamic metropolitan ring, whereas the central city, which
now accounts for 46% of the total population and a much
smaller proportion of built space, remains stagnant.
For the purpose of this study, two features of the metro-
politan space should be emphasized. The first is its huge land-
scape structural diversity. A number of sectors or areas of
particular significance can be identified in the metropolitan
territory, from the point of view of the physical medium as
well as from that of its humanization. This study focuses on
two sectors. One is the Guadalquivir Valley, which encom-
passes the river’s flood plain and its low terraces. This sector,
which is especially valuable for agricultural use, is home to
the central city and a traditional network of large, rural settle-
ments. The other is the Aljarafe plateau dating from the
Miocene Epoch, which rises one hundred meters above the
valley in the western part of the metropolitan territory. Land
occupation here is more fragmented, with a network of denser
yet smaller settlements and more favorable microclimatic
conditions than on the valley floor, which has made it more
attractive for suburban residential developments.
The metropolitan area’s second feature is the substantial
fragmentation of its territorial administrative organization, cov-
ering 46 different municipalities (Figure 1). Up until 2009, a
lack of planning tools at a metropolitan level has meant that the
size, shape, and intensity of urban growth in the area have been
subject to individualized designs with no coordination among
them, that is, via each of the 46 municipalities’ respective
Master Plans. This has led to mismatches and oversized urban
proposals and developments in the planning design of the met-
ropolitan space as a whole.
The base cartography used in the landscape analysis was the
Andalusian Land Use/Land Cover Map for 1984, 1999, and
2007 on a scale of 1:25,000, prepared by the Regional
Government’s Department for Environment. A ten-meter
spatial resolution for rasterization was adopted as it retains
an adequate level of detail of the landscape mosaic patches
while also minimizing any loss of information from the orig-
inal vectorial map due to the elimination of small patches or
linear landscape elements.
In terms of thematic resolution, a dual approach was
adopted with a low-resolution classification assigned as
Level 1 (four land use classes) and a medium resolution or
Level 2 classification (fourteen classes; see Table 1). Level 1
allows for a comprehensive analysis of the landscape pattern,
making it possible to immediately identify the main struc-
tural components and their spatial configuration. Meanwhile,
Level 2 adds greater detail to the first analysis, enabling spe-
cific structural patterns to be recognized and providing a
more accurate environmental-territorial interpretation of the
Following a two-scale methodological approach, analysis
was conducted on the metropolitan area as a whole and on
Figure 1. The metropolitan area of Seville.
Source: Prepared by authors.
Note: The figure on the right shows the 46 municipalities in the metropolitan area.
4 Journal of Planning Education and Research 00(0)
two different landscape units. The delimitation criteria for
both landscape units were adapted from those established by
Li et al. (2001). They were (1) a clear and intuitive differen-
tiation between the units; (2) consideration of the geomor-
phology, type, and intensity of urban development, and type
and level of natural resource use in each part of the territory;
and (3) delimitation that guarantees the integrity of the main
landscape mosaic patches.
Figure 2 shows the results of the delimitation process. As
previously mentioned, two of the most representative land-
scape units in the metropolitan area were selected for analy-
sis; we labeled them Unit A and Unit B. Unit A covers the
Guadalquivir river’s alluvial corridor from the metropolitan
area’s northern boundary to the marshlands that dominate the
southern part of the area. This includes the city of Seville, the
central urban hub of the metropolitan area. Unit B encom-
passes the westernmost sector of the metropolitan area and
comprises the “Aljarafe” residential zone and the central sec-
tor along the Guadiamar river basin, a major regional eco-
Recommendations made by a range of authors were taken
into account with regard to the core criteria for selecting
landscape metrics and the proposals for specific metric sets
(Botequilha-Leitão and Ahern 2002; Cushman, McGarigal,
and Neel 2008; Neel, McGarigal, and Cushman 2004). A set
of seven landscape metrics was selected, corresponding to a
class-level analysis. These metrics were calculated for the
metropolitan area and for units A and B using FRAGSTATS
v.4.1 (McGarigal, Cushman, and Ene 2012). Subsequently, a
derived metric, which was the result of combining two other
indices (largest patch index/percentage of landscape [LPI/
PLAND]), was also included. The complete set of metrics is
shown in Table 2. Taken together, the selected metrics
enabled a synthetic characterization of the metropolitan
landscape’s composition and structure, providing relatively
easy-to-interpret measures of dominance, fragmentation,
isolation, and connectivity for the different land uses under
consideration. The information yielded by the different indi-
ces proved relevant in relation to artificial surfaces—for ana-
lyzing urban growth patterns—and in relation to open or
nonbuilt space—for analyzing the metropolitan landscape’s
Land use change matrices for Level 1 classification were
also calculated to complement the landscape metrics analy-
sis. They covered the 1984–1999 and 1999–2007 periods for
the metropolitan area and for units A and B. Landscape met-
rics tables and land use change matrices are included in the
Online Supplemental Appendix.
A continuous expansion of the artificial uses class was
observed at a metropolitan level across the entire study
period, with a slight increase in growth rate between 1999
and 2007 (see Figures 3 and 4). In addition, the average size
(AREA_MN) of the artificial class patches grew steadily
over the study period. In contrast, the number of patches
(NP) fell between 1999 and 2007 (see Figure 5). This points
to a conurbation process or the merging of initially separate
urban nuclei as a result of their respective urban expansions.
A large proportion of artificial uses growth was concentrated
in the largest patch, which represents the central city, across
the entire study period. In 2007, this central patch repre-
sented almost 60% of the total surface area corresponding to
artificial uses (LPI/PLAND = 58.59%), thus indicating the
consolidation of a highly centralized urban pattern.
Unit A reveals a similarly monocentric spatial pattern for
artificial uses as well as a clear tendency to conurbation, as
reflected in a progressive reduction in NP and an increase in
AREA_MN (see Figure 5). Consequently, in 2007, the cen-
tral patch represented 78% of the unit’s surface area devoted
to artificial uses. Although artificial uses had a greater pres-
ence in Unit A, where the central city is located, the urban
growth process was far more intense in Unit B, reporting a
noticeable rate increase between 1999 and 2007 (see Figure
4). The largest artificial class patch in this unit grew between
1984 and 2007, representing over 33% of the class’ overall
surface area by the end of the study period (LPI/PLAND =
33.66%). This indicates the consolidation of a secondary
urban center below the metropolitan level. Despite the results
pointing to lesser fragmentation of artificial uses in Unit B,
on the whole, they continue to exhibit a more polycentric
structure and a more dispersed expansion dynamic than in
Unit A (lower AREA_MN values and a continuous increase
in NP throughout the study period).
Table 1. Land Use Classification.
Level 1 Level 2
Artificial surfaces Continuous urban fabric
Discontinuous urban fabrica
Other artificial surfacesb
Urban green areas
Water bodies/wetlands Water bodies/wetlands
Agricultural areas Irrigated agricultural areas
Nonirrigated agricultural areas
Heterogeneous agricultural areasc
Natural areas Forest/woodland
aLow-density residential developments, agriresidential developments.
bIndustrial, commercial, and transport units, mining sites, construction sites.
cMixed cultivation patterns.
dBare rock, burnt areas, and so on.
Feria Toribio and Santiago Ramos 5
Figure 3 compares the Level 1 land use classes of the
Metro area, Unit A, and Unit B for the time periods under
study, showing clear differences in the spatial pattern of arti-
ficial uses and its evolution. Furthermore, Level 2 metrics
underline the differences between both landscape units,
revealing more intense growth across all artificial land use
classes in Unit B. The growth of the discontinuous urban fab-
ric and the “other artificial surfaces” class in Unit B was, to
a greater extent, articulated around the generation of new
patches, especially during the 1999–2007 period. The largest
patch of continuous urban fabric holds greater importance in
Unit A than in Unit B, which heightens the contrast between
the aforementioned monocentric and polycentric urban
Natural cover experienced a continuous decrease across the
whole metropolitan area during the study period, although its
loss was more pronounced between 1999 and 2007. This
decrease reached 77.90% due to cropland conversion
(6,689.00 ha) and 18.81% due to artificial use transformation
Figure 2. Delimitation of landscape units based on three factors: (A) relief, (B) lithological units, and (C) land use/land cover.
Source: Prepared by authors with data provided by the Department for Environment, Regional Government of Andalusia.
Table 2. Landscape Metrics.
Landscape feature Landscape metrics Abbreviation
Dominance Percentage of landscape PLAND
Largest patch index LPI
Largest patch index/percentage of landscape LPI/PLAND
Fragmentation Number of patches NP
Mean patch area AREA_MN
Mean patch area standard deviation AREA_SD
Isolation Mean Euclidean nearest neighbor ENN_MN
Connectivity Area-weighted radius of gyration RGYR_AM
6 Journal of Planning Education and Research 00(0)
(1,614.81 ha) during the 1984–1999 period. In contrast, dur-
ing the 1999–2007 period, 55.49% of natural cover loss was
linked to the advance of agriculture (3,039.00 ha) and 42% to
the expansion of artificial uses (2,300.65 ha). In broad terms,
the metrics show natural cover to be the most fragmented
type of use, yielding a large number of patches with small
surface areas. The degree of fragmentation increased between
1999 and 2007 (see Figure 6), a process which was accompa-
nied by a strong loss of connectivity (see decreased RGYR_
AM in Figure 7). Furthermore, the largest patch fell from
over half of the natural surface cover in 1999 to only a third
Regarding the natural cover’s internal composition (Level
2 classification), a dominance of forest/woodland is observed
throughout the whole study period, followed by scrub/herba-
ceous vegetation cover. The results show an ongoing loss of
forest/woodland habitat, which was somewhat more pro-
nounced by the end of the study period, and a decrease in
scrub/herbaceous vegetation areas between 1999 and 2007.
A trend toward greater fragmentation of both land cover
typologies could be observed during the latter period. The
metrics also point to a considerable loss of connectivity
(RGYR_AM) for wooded cover between 1999 and 2007.
The results reveal significant differences between the
metropolitan area and the two landscape units under analy-
sis. In Unit A, the loss of natural habitat reached 61.39%
(494.85 ha) due to the advance of artificial uses during the
1984–1999 period, rising to 72.66% (723.59 ha) between
Figure 3. Evolution of land use in the metropolitan area and in units A and B.
Source: Prepared by authors with data provided by the Department for Environment, Regional Government of Andalusia.
Feria Toribio and Santiago Ramos 7
Figure 4. Dominance metrics (Level 1 land use classification).
Source: Prepared by authors.
Note: LPI (%) and PLAND (%) values for 1999 and 2007 are expressed as a percentage of the 1984 values (equal to 100%). The evolution of the
dominance metrics shows significant differences between the study units in relation to the percentage of land (PLAND) covered by each class and the
relative importance of the largest patch corresponding to each class (LPI).
Figure 5. Fragmentation metrics (artificial surfaces).
Source: Prepared by authors.
Note: AREA_MN (ha) and NP (dimensionless) values for 1999 and 2007 are expressed as a percentage of the 1984 values (equal to 100%). Changes to the
number of patches (NP) corresponding to artificial land uses and their mean area (AREA_MN) point to a greater fragmentation of the urban fabric in Unit
B, especially with regard to the “discontinuous urban fabric” and “other artificial surfaces” land use classes.
8 Journal of Planning Education and Research 00(0)
Figure 6. Fragmentation metrics (natural areas).
Source: Prepared by authors.
Note: AREA_MN (ha) and NP (dimensionless) values for 1999 and 2007 are expressed as a percentage of the 1984 values (equal to 100%). Although a
general tendency to fragmentation was observed for natural cover across the whole metropolitan area, the changes in fragmentation metrics revealed a
high degree of variation among the study units considered.
Figure 7. Connectivity and isolation metrics (natural areas).
Source: Prepared by authors.
Note: ENN_MN (m) and RGYR_AM (m) values for 1999 and 2007 are expressed as a percentage of the 1984 values (equal to 100%). A considerable loss
of connectivity (RGYR_AM) and a moderate increase in the distance between patches (ENN_MN) of natural cover can be observed at a metropolitan
scale of analysis. These tendencies vary considerably when units A and B and Level 2 classes are considered.
Feria Toribio and Santiago Ramos 9
1999 and 2007. The natural cover’s degree of fragmentation
was greater in this unit, yielding a high number of small-
sized patches. Scrub/herbaceous vegetation cover was the
majority natural class in Unit A. Its connectivity grew
between 1999 and 2007, as did the relative importance of its
largest patch. However, the configuration of forest/woodland
cover in this unit was very fragmented, with a loss in con-
nectivity between 1999 and 2007.
In Unit B, where natural cover is more abundant, losses
corresponding to this class due to the advance of artificial
uses grew from 7.98% (299.50 ha) in the 1984–1999 period
to 45.41% (485.41 ha) in the 1999–2007 period. The spatial
configuration of natural areas exhibited a dual nature and
was characterized by the presence of a main reservoir that
represented almost 50% of the class’ surface area (LPI/
PLAND = 49.44%) for the whole study period, plus a large
number of smaller patches. A fragmentation process (see
Figure 6) and the growth of the largest patch between 1999
and 2007 heightened this duality. The forest/woodland cover
was the majority natural class in this unit. A constant decrease
was observed for this class, which is reflected in a notable
loss of connectivity and a reduction in the largest patch.
Meanwhile, a spread of scrub/herbaceous vegetation cover
was observed, yielding a reduction in its degree of fragmen-
tation and an increase in connectivity throughout the study
Despite being the majority use, a moderate reduction in crop-
land surface area was observed throughout the study period.
Croplands were lost mainly through their conversion to arti-
ficial uses, especially between 1999 and 2007, when this pro-
cess represented 68.61% of the total agricultural cover loss
(6,209.93 ha). Most agricultural use corresponded to nonir-
rigated agricultural areas, which, despite their continuous
decrease, always maintained a presence of over 40% of the
landscape. Although their total surface area was smaller, the
irrigated agricultural areas expanded noticeably throughout
the whole study period.
In Unit A, a fall in agricultural surfaces was linked to the
loss of nonirrigated areas. In contrast, irrigated crops were
the dominant agricultural use and experienced net growth
throughout the study period. The occupation of croplands by
artificial land uses represented 76.32% of cropland loss
between 1999 and 2007 (1,470.05 ha).
In Unit B, agricultural surfaces remained relatively stable
between 1984 and 1999, until experiencing a decrease
between 1999 and 2007. The pressure from artificial uses on
crops was even more noticeable in Unit B than in Unit A,
representing 80.42% of cropland loss between 1999 and
2007 (2,960.56 ha). Although the dominance of nonirrigated
agricultural areas in this unit is clear, the relative growth of
irrigated areas is worth highlighting, exceeding 100% across
the entire study period.
Urbanization: Different Growth Patterns within
the Same Metropolitan Area
As a whole, the results show that the Seville Metropolitan
Area has experienced a continuous expansion process
throughout the entire study period, with a greater growth rate
between 1999 and 2007. The growth in artificial uses mainly
takes place around the metropolitan area’s central nucleus,
which points to the consolidation of the area’s monocentric
pattern. This trend is reinforced by the conurbation processes
The spread of artificial uses between 1999 and 2007 is
particularly linked to the greater dynamism of the infrastruc-
ture and large centers of activity (commercial, industrial,
etc.) that comprise the “other artificial surfaces” category,
and less so to residential uses. As such, the urban growth
process during this period is highly conditioned by the more
fragmented and discontinuous spatial configuration of these
land use typologies. Any attempts at controlling urban expan-
sion should recognize the dynamic nature of nonresidential
uses and incorporate specific measures to regulate them. The
dispersed expansion of commercial and industrial uses is
most likely due to the lack of a coordinated planning strat-
egy, thus leading to competition among the different munici-
palities. It is especially relevant given that strategies aimed at
preventing urban sprawl tend to focus primarily on the regu-
lation of residential uses.
In terms of the comparison between the metropolitan area
and the two landscape units under study, the trend graphs
show common overall urban expansion patterns for the area
as a whole and for units A and B. However, the different rates
of change, the evolution of the various types of artificial use,
and, above all, the contrast in the absolute values displayed
by the metrics enable a different model of spatial organiza-
tion to be identified for each unit. In very general terms, Unit
A corresponds to the global evolution of the metropolitan
area, monocentric in character due to the presence of the cen-
tral city. In contrast, growth is more intense in Unit B during
the most recent period and shows a more fragmentary con-
figuration, with less relative importance given to its central
To summarize, the two-scale analysis shows that the char-
acter of urban growth patterns in the metropolitan area is
scale dependent. Although the metropolitan area as a whole
can be considered morphologically monocentric, with a
powerful core that tends to strengthen its role, in Unit B, the
urban growth pattern is polycentric, more dynamic, and leans
toward a fuzzier spatial distribution. This could signal a less
sustainable urban configuration in terms of mobility and land
consumption. Consequently, planning measures should
respond differently to both urban realities. From this per-
spective, reinforcing a polycentric urban system articulated
around a defined set of compact, mid-density urban areas
10 Journal of Planning Education and Research 00(0)
would be an advisable option for Unit B. For this purpose,
the design of a metropolitan-scale green infrastructure can
become a key planning tool. The notion of a green infrastruc-
ture involves the strategic planning of a network of natural
and agricultural areas that possess high environmental value
and has proved to be a useful tool for containing urban sprawl
processes (Benedict and MacMahon 2002; European
Environment Agency 2011). In this context, the Metropolitan
Plan—which acts as a coordination instrument for the Master
Plans of the different municipalities—is the most appropriate
way to articulate a green infrastructure at this scale as it
would allow for open space planning and urban growth regu-
lation measures to be integrated into a single comprehensive
strategy. An alternative would be to develop a specific green
infrastructure plan in the area. However, given the ever-
increasing pressure that urban development places on natural
habitats, its complementarity with urban growth regulation
measures is indispensable to achieve success.
Natural Cover: The Basis for a Metropolitan-
Scale Green Infrastructure
Natural habitats undoubtedly represent the backbone of a met-
ropolitan-scale green infrastructure proposal. And natural
cover in the metropolitan area presents a clearly dual configu-
ration. On one hand, a considerable proportion is concentrated
around large patches in peripheral locations, which act as natu-
ral reservoirs on a metropolitan scale. On the other hand, there
are vast quantities of smaller-sized patches that can play an
important role when it comes to carrying out ecological func-
tions within an agricultural or urban matrix. However, they
may see their value limited and may become more vulnerable
to urban encroachments because of their high degree of frag-
mentation. Four key issues should be highlighted. First, a
greater decrease in natural cover can be noticed between 1999
and 2007. Second, an increase in the fragmentation of the nat-
ural class and a heavy loss of connectivity were observed dur-
ing this latter period, with both processes exerting a greater
impact on forest/woodland than on the non–forest/woodland
habitat. Third, there was a sharp increase in pressure from arti-
ficial uses on the natural class between 1999 and 2007. And
fourth, in general terms, the impact of artificial uses on the
spatial configuration of natural habitats seems to be just as rel-
evant as the net loss of natural surfaces.
Clear-cut differences can be discerned between the find-
ings for units A and B and for the metropolitan area as a
whole, especially in relation to the loss of natural cover—
which is particularly intense in Unit A—and the increase in
connectivity—which can be observed for both units. These
local processes, which are undoubtedly relevant at a submet-
ropolitan scale, would have remained hidden had the analy-
sis been restricted to the urban area as a whole.
Based on these results, the design of a metropolitan-scale
green infrastructure should be able to integrate both large,
peripheral natural areas and the smaller, close to urban natu-
ral patches into a single interconnected system. Additionally,
this green system should respond to the different structural
characteristics and processes detected for the natural cover in
The need to preserve the most fragmented habitats from
urban growth in Unit A has been identified, with commitment
to a green belt strategy as a possible response to maintaining
the continuity of the whole ambit and to strengthening the
habitats’ environmental and public use functions, which may
boast strategic value given their proximity to the central city.
Some basic planning guidelines would be the following:
The natural areas to the south of the central urban
nucleus currently constitute a green belt. It is essential
to guarantee their preservation against possible future
urbanization processes and to maintain their continu-
In contrast, the development of a greening strategy in
the northern sector is recommended. This encom-
passes both the design of key green spaces and the
establishment of corridors that ensure spatial integrity
of the green system. Coordination between the sec-
tor’s municipalities is necessary for designing an
infrastructure that responds to a systemic, supramu-
Any green planning strategy should consider the
Guadalquivir river as the essential green corridor of
this unit. It connects the green areas located within the
urban fabric with the natural areas north and south of
the urban nucleus. Similarly, it serves as an axis that
connects the whole central urbanized area with the
peripheral natural spaces of Unit A and of other units
to its north and south.
A mixed approach is necessary in Unit B to preserve the
extensive peripheral habitats from the advance of agricul-
ture, especially, in those areas where wooded natural cover is
falling. Simultaneously, smaller habitats need to be protected
from increasing urban pressure. From a planning perspec-
tive, it is crucial to coordinate the area’s multiple municipali-
ties in a concerted effort in order to preserve the few natural
areas that remain in this rapidly urbanizing environment and
to design green corridors that give spatial and functional con-
tinuity to the system. In a context where periurban natural
cover is particularly scarce, it is also important to address the
potential of agricultural areas when it comes to constructing
a green system.
Agricultural Areas: A Strategic, Multifunctional
Resource for Metropolitan Planning
The results show the key role played by agricultural areas in
shaping the metropolitan land mosaic. Cropland represents
Feria Toribio and Santiago Ramos 11
the predominant land use in the metropolitan area, covering
over 70% of its surface; it constitutes the landscape matrix
and has a major presence on the periurban fringe. It also
adopts a dual role in the dynamics of metropolitan change:
passively, as a primary support to the growth of artificial uses
and actively, as an important vector of natural habitat surface
loss. For these reasons, and coupled with all its economic
and environmental value, the agricultural matrix takes on an
importance that should be fully acknowledged in the context
of planning processes. What is more, it should not be viewed
as a subsidiary component, which is how it is usually consid-
ered in local planning experiences.
In Unit A, irrigated crops (mainly rice crops and fruit tree
groves) make up the dominant land use. Their economic
profitability is clearly greater than that of nonirrigated crops,
a fact that gives them a certain intrinsic resistance to the
intense urbanization pressure observed at the core of the met-
ropolitan area. From an environmental point of view, irri-
gated crops perform an important flood protection function,
so any efforts to guarantee their preservation are highly
advisable. In Unit B, nonirrigated crops—mainly linked to
olive groves in this sector—are predominant. Centenary
olive tree plantations constitute an essential component of
the area’s traditional landscape, yielding an undeniable heri-
tage value. Their environmental value is also enhanced by a
lack of nearby natural areas; this means that they even func-
tion as the true urban forest of this metropolitan sector, pro-
viding ecosystem services such as atmospheric C
sequestration and microclimate regulation in periurban areas.
The analytic results clearly show that their resistance against
urban sprawl is lower than that of irrigated crops.
Given these outcomes, introducing periurban agriculture
as a multifunctional component into the metropolitan green
infrastructure is strongly recommended. Although urban
agriculture has traditionally played a secondary role in urban
and metropolitan planning initiatives, there are many world-
wide examples of successful green systems that benefit from
the integration of cropland areas (Lohrberg et al. 2015;
Lovell 2010; Zasada 2011), demonstrating how agriculture
can add value to green infrastructures and reinforce sprawl
control measures. As mentioned above, natural areas are par-
ticularly scarce in the northern part of Unit A and in the east-
ern sector of Unit B. In both cases, it is recommended that
the cultivated areas of greater environmental value be
included in the metropolitan green system to strengthen its
spatial integration and ensure its effectiveness at controlling
The two-scale methodological approach developed confirms
the hypothesis that the change processes analyzed are differ-
ent in character according to the metropolitan area’s structural
landscape diversity. Some change trends appear to be general,
affecting both the metropolitan area and the two landscape
units; however, the absolute values of the indices and the rates
of change show significant differences between both units. In
some cases, opposite processes are observed depending on
the area under analysis. We can draw the conclusion that, due
to its complex and diverse nature, the metropolitan territory is
not a homogeneous or isotropic environment where spatial
processes manifest uniformly and, for that reason, it cannot
be planned for without catering to this internal diversity.
Neither does it seem possible to talk about a clear urban–rural
gradient, despite the area’s marked monocentric character.
Rather, we find a combination of units with a specific internal
configuration as well as determinants that generate differenti-
ated regional/local dynamics.
Accordingly, not only is a disaggregated analysis by land-
scape unit strongly recommendable for landscape change
monitoring, but a multiscale planning approach also seems to
be useful to address metropolitan landscape complexity.
As previously mentioned, the approach could prove reli-
able when applied to other metropolitan areas with a similar
level of complexity relative to their internal spatial configu-
ration. Going back to the initially cited examples, in the case
of Copenhagen, the multiscale approach to landscape analy-
sis developed in this city can prove useful in identifying and
monitoring specific land use change dynamics and trends
across different zones of the green wedge system, for exam-
ple, the radial wedges between the fingers of the city, the
transverse wedges, and the wedges within the city’s fingers.
Its usefulness can also extend to evaluating the specific evo-
lution of each urban finger in relation to the surrounding
nonbuilt space to ensure balanced access to the green infra-
structure across all sectors of the metropolitan area. In the
case of Stuttgart, the multiscale approach could be useful for
monitoring whether the determinations outlined in the
Regional Plan are being upheld, especially considering the
general lack of evaluation processes of territorial plans by
the competent German authorities (Hildenbrand 2017).
Regarding the Barcelona Metropolitan Region, the method-
ology put forward in this paper could prove useful for the
design of the linear structural elements—for example, eco-
logical corridors—that come under the Regional Plan’s Open
Space System (Forman 2008) as well as for the monitoring
and evaluation of land use changes across the different terri-
torial zones defined in the Plan.
Finally, from the perspective of the case subject to this
research, some insights in relation to planning guidelines can
be provided for the metropolitan area of Seville, which could
also be applied to other metropolitan contexts:
1. Based on the results, it can be stated that while urban
growth is a structural, long-term process, in the met-
ropolitan area of Seville, it intensified, showing dis-
tinctive spatial patterns over the 1999–2007 period,
corresponding to a national real estate boom cycle. A
greater impact on the natural environment was
observed for this period. An important finding is that
12 Journal of Planning Education and Research 00(0)
changes to the natural habitat’s configuration (frag-
mentation, loss of connectivity) were just as relevant
as the net loss of the natural surface area itself.
Furthermore, these processes were not only linked to
the quantitative dimension of land use transformation
but also to its specific spatial display. Thus, urban
growth regulation policies at a metropolitan scale
should not be restricted to merely establishing the
dimensional thresholds for the city’s spatial expan-
sion, something which is commonplace in Spanish
urban and regional planning practices. It is perhaps
more important to consider the spatial configuration
of these growth processes: first, in relation to urban
land uses that display a dispersed, fragmentary con-
figuration and second, with regard to the develop-
ment of new transport infrastructures that have a
major impact on the natural habitat’s continuity.
Lastly, although urban growth processes have slowed
significantly in Spain since 2008 because of the eco-
nomic crisis, it is always deemed appropriate to con-
tinually monitor subsequent land use change
processes and to evaluate the impact of planning
2. The fragmentation and loss of connectivity of natural
habitats are processes that heavily affect the resil-
ience and ecological functionality of green spaces in
metropolitan areas. A suitable series of planning poli-
cies and measures that address the preservation,
enhancement, and spatial articulation of the metro-
politan open space system can help maintain the con-
tinuity of natural areas and the provision of ecosystem
services. To this end, it is essential to identify and
preserve potential elements that might improve con-
nectivity, particularly, those linked to the river net-
work. However, this also encompasses other linear
elements including rural roads, parkways, and aban-
doned railways. The development of a planning strat-
egy aimed at building an integrated metropolitan
green infrastructure could represent a major advance
in this respect, helping to increase habitat continuity
and preserving it from future urban expansion.
3. As has been pointed out in the context of both tradi-
tional regional planning theory (Luccarelli 1996) and
current European territorial policies (Territorial
Agenda of the European Union 2007), the relevant
role played by agricultural spaces in shaping the met-
ropolitan region should be properly recognized in
planning strategies. Not only should their ecological
value and environmental quality be considered, but
also their cultural heritage assets and their contribu-
tion to metropolitan economic activity. Reversing the
subsidiary role they frequently adopt, at least in
Spanish planning practices, the major role of agricul-
tural areas needs to be acknowledged not only on
analytical and normative levels but also when it
comes to providing financial support to, for example,
ecologically friendly practices or proximity agricul-
ture. This is recommended for two reasons: first, to
preserve and improve the proper functioning of eco-
logical systems and second, to maintain and enhance
the important economic and environmental benefits
that agricultural areas offer to the metropolitan city.
The authors would like to thank the anonymous reviewers whose
comments and suggestions helped improve and clarify this
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect
to the research, authorship, and/or publication of this article.
The author(s) disclosed receipt of the following financial support
for the research, authorship, and/or publication of this article: This
work has been supported by the Spanish Ministry of Economy and
Competitiveness; project CSO2014-55780-C3-1-P (National
R&D&I Plan). The funding source had no role in the study design;
in the collection, analysis, and interpretation of data; in the writing
of the report; or in the decision to submit the paper for publication.
Supplemental material for this article is available online.
Aguilera, Francisco, Luis M. Valenzuela, and André
Botequilha-Leitão. 2011. “Landscape Metrics in the
Analysis of Urban Land Use Patterns: A Case Study in
a Spanish Metropolitan Area.” Landscape and Urban
Planning 99 (3–4): 226–38.
Albertos, Juan Miguel, and José L. Sánchez, eds. 2014. Geografía
de la crisis económica en España [Geography of the economic
crisis in Spain]. Valencia: Publicaciones de la Universidad de
Angel, Shlomo, Jason Parent, and Daniel L. Civco. 2012. “The
Fragmentation of Urban Landscapes: Global Evidence of a
Key Attribute of the Spatial Structure of Cities 1990-2000.”
Environment and Urbanization 24 (1): 249–83.
Benedict, Mark, and Edward T. MacMahon. 2002. Green
Infrastructure: Smart Conservation for the 21st Century.
Washington, DC: Sprawl Watch Clearinghouse.
Botequilha-Leitão, André, and Jack Ahern. 2002. “Applying
Landscape Ecological Concepts and Metrics in Sustainable
Landscape Planning.” Landscape and Urban Planning 59:65–
Coskun, Hepcan C. 2013. “Quantifying Landscape Pattern and
Connectivity in a Mediterranean Coastal Settlement: The Case
of the Urla District, Turkey.” Environmental Monitoring and
Assessment 185 (1): 143–55.
Cushman, Samuel A, Kevin McGarigal, and Maile C. Neel. 2008.
“Parsimony in Landscape Metrics: Strength, Universality and
Consistency.” Ecological Indicators 8:691–703.
Feria Toribio and Santiago Ramos 13
Dewan, Ashraf, Yasushi Yamaguchi, and Md. Ziaur Rahman.
2012. “Dynamics of Land Use/Cover Changes and the
Analysis of Landscape Fragmentation in Dhaka Metropolitan,
Bangladesh.” Geojournal 77 (3): 315–30.
European Environment Agency. 2011. Green Infrastructure and
Territorial Cohesion: The Concept of Green Infrastructure
and Its Integration into Policies Using Monitoring Systems.
Copenhagen: European Environment Agency.
Feria, José María. 2018. “Crecimiento urbano, crisis inmobiliaria
y planificación metropolitana en España” [Urban growth, real
estate crisis and metropolitan planning in Spain]. Ciudad y ter-
ritorio. Estudios territoriales 198:651–69.
Feria, José María, and Lucas Martínez. 2016. “La definición y
delimitación del sistema metropolitano español: permanencias
y cambios entre 2001 y 2011” [The definition and delimitation
of the Spanish metropolitan system: Permanences and changes
between 2001 and 2011]. Ciudad y Territorio. Estudios
Feria, José María, and Jesús Santiago. 2009. “Funciones ecológicas
del espacio libre y planificación territorial en ámbitos metro-
politanos: perspectivas teóricas y experiencias recientes en el
contexto español” [Metropolitan spatial planning and the eco-
logical functions of open spaces: theoretical perspectives and
recent experiences in Spain]. Scripta Nova. Revista Electrónica
de Geografía y Ciencias Sociales 299 (XIII). http://www.ub.es/
Forman, Richard T. T. 2008. Urban Regions: Ecology and
Planning beyond the City. Cambridge: Cambridge University
Forman, Richard T. T., and Michel Godron. 1986. Landscape
Ecology. California: John Wiley & Sons.
Furberg, Dorothy, and Yifang Ban. 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): 1068–88.
Hall, Peter. 1998. Cities in Civilization: Culture, Technology, and
Urban Order. New York: Pantheon Books.
Hildenbrand, Andreas. 2017. Gobernanza y planificación
territorial en las áreas metropolitanas [Governance and spatial
planning in metropolitan areas]. Valencia: Publicaciones de la
Universidad de Valencia.
Johnson, Elizabeth A., and Michael W. Klemens. 2005. “The
Impacts of Sprawl on Biodiversity.” In Nature in Fragments:
The Legacy of Sprawl, edited by Elizabeth A. Johnson,
Michael W. Klemens, 18–53. New York: Columbia
Kaza, Nikhil. 2013. “The Changing Urban Landscape of the
Continental United States.” Landscape and Urban Planning
110 (1): 74–86.
La Rosa, Daniele, and Riccardo Privitera. 2013. “Characterization
of Non-Urbanized Areas for Land-Use Planning of Agricultural
and Green Infrastructure in Urban Contexts.” Landscape and
Urban Planning 109 (1): 94–106.
Li, Xin, Ling Lu, Guodong Cheng, and Honglang Xiao. 2001.
“Quantifying Landscape Structure of the Heihe River Basin,
North-West China, Using FRAGSTATS.” Journal of Arid
Lizée, Marie-Hélène, Stéphanie Manel, Jean-François Mauffrey,
Thierry Tatoni, and Magali Deschamps-Cottin. 2012. “Matrix
Configuration and Patch Isolation Influences Override the
Species–Area Relationship for Urban Butterfly Communities.”
Landscape Ecology 27 (2): 159–69.
Lohrberg, Frank, Lilli Lička, Lionella Scazzosi, and Axel Timpe,
eds. 2015. Urban Agriculture Europe. Berlín: Jovis Verlag
Lovell, Sarah Taylor. 2010. “Multifunctional Urban Agriculture
for Sustainable Land Use Planning in the United States.”
Sustainability 2 (8): 2499–2522.
Luccarelli, Mark. 1996. Lewis Mumford and the Ecological Region:
The Politics of Planning. Cambridge: Cambridge University
Lynch, Amy J. 2016. “Is It Good to Be Green? Assessing the
Ecological Results of County Green Infrastructure Planning.”
Journal of Planning Education and Research 36 (1): 90–104.
Magle, Seth, Kristin A. Salamack, Kevin R. Crooks, and Richard P.
Reading. 2012. “Effects of Habitat Fragmentation and Black-
Tailed Prairie Dogs on Urban Avian Diversity.” Biodiversity
and Conservation 21 (11): 2803–21.
Marzluff, John M., and Kern Ewing. 2008. “Restoration of
Fragmented Landscapes for the Conservation of Birds: A
General Framework and Specific Recommendations for
Urbanizing Landscapes.” In Urban Ecology, edited by John
M. Marzluff, Eric Shulenberger, Wilfried Endlicher, Marina
Alberti, Gordon Bradley, Clare Ryan, Ute Simon, and Craig
ZumBrunnen, 739–56. New York: Springer.
McGarigal, Kevin., Sam A. Cushman, and Eduard Ene. 2012.
FRAGSTATS V4: Spatial Pattern Analysis Program for
Categorical and Continuous Maps. Amherst: University of
Neel, Maile C., Kevin McGarigal, and Samuel A. Cushman.
2004. “Behavior of Class-Level Landscape Metrics across
Gradients of Class Aggregation and Area.” Landscape
Organisation for Economic Co-operation and Development. 2012.
Redefining “Urban”: A New Way to Measure Metropolitan
Areas. Paris: Organisation for Economic Co-operation and
Parrish, Michael C., and Jeffrey Hepinstall-Cymerman. 2012.
“Associations between Multiscale Landscape Characteristics
and Breeding Bird Abundance and Diversity across Urban-
Rural Gradients in Northeastern Georgia, USA.” Urban
Ecosystems 15 (3): 559–80.
Pijanowski, Bryan C., and Kimberly D. Robinson. 2011. “Rates
and Patterns of Land Use Change in the Upper Great Lakes
States, USA: A Framework for Spatial Temporal Analysis.”
Landscape and Urban Planning 102 (2): 102–16.
Reice, Seth R. 2005. “Ecosystem, Disturbance and the Impact of
Sprawl.” In Nature in Fragments: The Legacy of Sprawl, edited
by Elizabeth A. Johnson and Michael W. Klemens, 90–108.
New York: Columbia University Press.
Shrestha, Milan, Abigail Mara York, Christopher G. Boone, and
Sainan Zhang. 2012. “Land Fragmentation Due to Rapid
Urbanization in the Phoenix Metropolitan Area: Analyzing the
Spatiotemporal Patterns and Drivers.” Applied Geography 32
Territorial Agenda of the European Union. 2007. “Territorial Agenda
of the European Union: Towards a More Competitive and
Sustainable Europe of Diverse Regions.” Agreed at the Occasion
of the Informal Ministerial Meeting on Urban Development and
Territorial Cohesion, May 24–25. http://ec.europa.eu/regional_
14 Journal of Planning Education and Research 00(0)
Tian, Guangjin, Jing Jiang, Zhifeng Yang, and Yaoqi Zhang. 2011.
“The Urban Growth, Size Distribution and Spatio-Temporal
Dynamic Pattern of the Yangtze River Delta Megalopolitan
Region, China.” Ecological Modelling 222 (3): 865–78.
United Nations. 2014. World Urbanization Prospects: The 2014
Revision. New York: United Nations, Department of Economic
and Social Affairs, Population Division. http://esa.un.org/unpd/
Wu, Jianguo, G. Darrel Jenerette, Alexander Buyantuyev, and
Charles Redman. 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.
Yang, Qi., Jianlong Li, Xiaoyu Gan, Jie Zhang, Feng Yang, and
Yurong Qian. 2012. “Comparison of Landscape Patterns between
Metropolises and Small-Sized Cities: A Gradient Analysis with
Changing Grain Size in Shanghai and Zhangjiagang, China.”
International Journal of Remote Sensing 33 (5): 1446–64.
Yin, Robert K. 2002. Case Study Research: Design and Methods.
Thousand Oaks: SAGE.
Zasada, Ingo. 2011. “Multifunctional Peri-Urban Agriculture. A
Review of Societal Demands and the Provision of Goods and
Services by Farming.” Land Use Policy 28 (4): 639–48.
José María Feria Toribio is a tenure professor in the Department
of Geography, History and Philosophy at Universidad Pablo de
Olavide, Seville (Spain). His research interests include urban and
regional planning, socioeconomic processes in metropolitan areas,
urban mobility, and housing.
Jesús Santiago Ramos is an assistant professor in the
Department of Geography, History and Philosophy at
Universidad Pablo de Olavide, Seville (Spain). His research
interests include urban and metropolitan green infrastructures,
urban and regional planning, and the analysis of ecosystem
services in urban areas.