Definition of homogenous environmental management units for the Catalan Coast
Brenner et al. 2006, Environmental Management
Environ Manage (2006) 38:993–1005
DOI 10.1007/s00267-005-0210-6
Definition of Homogeneous Environmental Management Units
for the Catalan Coast
Jorge Brenner Æ Jose A. Jimenez Æ Rafael Sarda
´ ´
Received: 12 July 2005 / Accepted: 29 May 2006
Ó Springer Science+Business Media, Inc. 2006
Abstract Geographical areas constitute the basic Keywords Environmental regionalization Æ
implementation locus for integrated coastal zone man- Homogeneous Environmental Management Units
agement strategies and activities. Because the definition (HEMUs) Æ Integrated Coastal Zone Management
of territorial planning objectives may be affected by (ICZM) Æ Geographic Information System (GIS)
socioeconomic and environmental characteristics, one
of the main steps in the process involves dividing the
coast into homogeneous environmental management Introduction
units (HEMUs). This article presents a general and
simple method for regionalizing the landside of a Management of coastal areas under the sustainable
coastal zone into HEMUs and illustrates it through regional development mandate is a complex process.
application to the Catalan coast. Socioeconomic and Difficulties arise from the need to strike a balance
natural (biophysical) subsystems were selected as the between socioeconomic development and coastal con-
most appropriate dimensions of the regionalization servation. This balance may vary due to the high var-
process. Dimensions were described using 11 spatial iability of the primary components of the coastal
themes, which were managed in a geographic informa- system, i.e., the natural and socioeconomic subsystems
tion system environment that proved to be an adequate (Van der Weide 1993). The aim of integrated coastal
tool for the purpose. A final coastal zone map of four management is to maintain a sustainable relationship
classes of HEMUs connected to local administrative between the resources of these two subsystems and
units was obtained, and because it reflects the current their exploitation, preventing (or mitigating) potential
natural and socioeconomic dynamics, it can be consid- conflicts and reducing the uncertainties associated with
ered as an initial step in the planning process for the planning and decision making. However, to manage a
Catalan coast. Although the proposed method was coastal region properly, a clear picture should already
developed based on the characteristics of the Catalan have been obtained of the expectations of stakeholders
coast, it is general enough to be adapted and applied to and/or society regarding each specific unit of territory,
most developed or developing coastal areas. as well as the legal framework into which it fits and the
existing property rights (Mee 2005). When this vision is
shared and accepted, specific criteria can be developed
J. Brenner (&) Æ J. A. Jimenez to accommodate uses of coastal areas, to resolve po-
´
Laboratori d’Enginyeria Marıtima, ETSECCPB,
tential conflicts, and to facilitate the decision-making
`
Universitat Politecnica de Catalunya,
process. In Spain, the coastal zone is administratively
`
C/ Jordi Girona, 1-3, Modul D-1, 08034, Barcelona, Spain
defined in the Coastal Law (BOE 1989) in terms of a
E-mail: jorge.brenner@upc.edu
marine and terrestrial zone that falls within the public
´
R. Sarda
domain. It is a very narrow fringe of territory delimited
Centre d’Estudis Avancats de Blanes, CSIC
¸
on the land side by the innermost high-water level.
´
Carrer d’Acces a la Cala Sant Francesc, 14, 17300
Inland there is a conservation easement fringe of
Blanes, Spain
123
994 Environ Manage (2006) 38:993–1005
human activity. Several analytical approaches have
variable width with different restrictions. Although this
been used, such as multivariate classifications/cluster-
implies some kind of management or regulation of
ing, factor analysis, fuzzy logic, multicriteria analysis,
activities, there is an overlap with the responsibilities
and spatial overlapping (Fricker and Forbes 1988,
of the regional and local administrations. These factors
Gornitz 1990, Bartley and others 2001, Baja and others
generate a relatively poorly defined area in terms of
2002, Escofet 2002, Maxwell and Buddemeier 2002,
planning and management.
Henocque and Andral 2003, Vafeidis and others 2004,
Integrated coastal zone management (ICZM) is a
´˜
Yanez-Arancibia and Day 2004). The most common
tool to help achieve sustainable regional development
HEMU definitions have been based on biophysical
in coastal areas. The main purpose of all ICZM ini-
characteristics such as geomorphology, climate, vege-
tiatives is to maintain, restore, or improve specific as-
tation, and biodiversity. However, in order to develop
pects of coastal zone systems and their associated
an integrated vision of the coastal zone, the socioeco-
human societies. An important feature of ICZM ini-
nomic dimension needs to be incorporated into the
tiatives is that they address the needs of both socio-
´
process (Sarda and others 2005).
economic development and natural conservation in
As a starting point, typologies constitute repeatable
geographically specific planning activities at multiple
homogeneous units that are the basis for division or
administrative levels. Thus, geographic areas constitute
classification into geographical units. Usually, devel-
the basic implementation locus of ICZM strategies and
opment of a typology for geospatial data takes either a
activities. Many authors have emphasized the role of
top-down or a bottom-up approach (Maxwell and
appropriate territorial information and organized,
Buddemeier 2002). The top-down approach to classi-
coherent databases as essential for decision making in
fication is based on a decision tree containing prede-
the coastal zone (e.g., Shupeng 1988, Bartlett 2000).
fined environmental characteristics that is specifically
The coastal zone is characterized by a high degree of
developed for a given environment (e.g., Finkl 2004).
natural and socioeconomic heterogeneity because of
In the bottom-up approach, a clustering method is used
the existence of multiple resources and uses, and its
to identify groups with similar environmental charac-
highly dynamic nature (McLaughlin and others 2002).
teristics. A variation on bottom-up classification is the
The spatial heterogeneity of the coastal zone can be
regionalization approach, which locates spatially con-
rationalized by selecting homogeneous environmental
tiguous class members after clustering without atten-
management units (HEMUs), discrete homogeneous
tion to spatial location (Harff and Davis 1990).
areas, or units with similar characteristics (for a
Regions, which constitute a unique discrete system,
description of similar approaches, see Christian 1958,
become planning units and can be identified by a spe-
Amir 1987, UNESCO 1997). These territorial units
cific valuable quantifiable phenomenon. A combina-
should then be linked to a strategic territorial plan, and
tion of structural and functional typologies can
thus, to active management units (Mee 2005). These
determine the specific processes that constitute indi-
units form the basis for research and data collection,
vidual management regions. A review of the biophys-
and subsequently become the boundaries defining
ical characteristics used for the classification of coastal
areas with similar land attributes selected as decision
and marine environments can be found in Finkl (2004).
criteria for planning and evaluation (Baja and others
The analytical process leading to regionalization can be
2002). This process of reducing spatial complexity is a
divided into two discretization strategies: hierarchical
way of linking management decisions to the biophysi-
´˜
unit grouping and segmenting (Yanez-Arancibia and
cal and socioeconomic properties of a territory, and
Day 2004). These two approaches tend to give rise to
thus, meets the need of policy makers to access quan-
regions based on a hierarchical criterion of belonging
titative information on physical areas. To be an effi-
to a higher scale unit; thus, units can be identified as
cient management tool, these should also be integrated
either belonging to a higher region or forming one
within the existing administrative framework. To
(Escofet 2002). The interactions between the individ-
properly define a HEMU, natural and socioeconomic
ual regions should determine the territorial planning
properties must represent the coastal system as closely
schema that management needs for the process of
as possible (Zonneveld 1994), and if they are imple-
reconciling the natural and socioeconomic subsystems.
mented in a geospatial management framework, all the
Because of the difficulties associated with this data-
elements of the system (natural, socioeconomic,
driven process, most planning instruments, such as
administrative, etc.) must be spatially coherent.
assessment and evaluation, lack this framework.
The definition of HEMUs is a common task
As a useful working concept, the ecosystem
when one is dealing with systems with different
approach has been defined to help in the process of
environmental properties that support significant
123
Environ Manage (2006) 38:993–1005 995
lem (MAUP) being one of the most notable. MAUP
setting environmental management boundaries (CBD
appears in spatially averaged studies when units are
1999). Large-scale applications of this approach can be
subdivided into smaller nonoverlapping units such that
found in the different global proposals for environ-
intrinsic geographical meaning is absent (Openshaw
mental regionalization, such as the large marine eco-
1984). It has major implications in two areas: (1) the
systems of the world (Sherman and Alexander 1986)
number of aerial divisions of a unit that can be per-
and the environmental land units of the European
formed, and (2) the data aggregation at different
ecological regions proposals (EEA 2003). This ap-
resolutions (Bian 1997, Cao and Lam 1997). Although
proach recognizes the dynamics and complexity of
several solutions have been proposed, the main
ecosystems in order to provide an analytical frame-
uncertainty arises when geospatial data are scale
work for the development of managerial strategies
dependent (Cao and Lam 1997, Marceau 1999). Its
(Rappaport 1999). An example of regionalization is the
importance increases with increasing spatial and tem-
use of river basins to define management units for use
´˜ poral heterogeneity of the coast, and the difficulty of
in a variety of approaches (e.g., Yanez-Arancibia and
combining natural and socioeconomic subsystem indi-
Day 2004). This approach is used in the European
cators in the assessment process further complicates
Union (EU) to apply the European Directive on Water
the final situation. Consequently, methodological dif-
Policy (EC 2000). Although this approach is logical for
ficulties are presented for the implementation of
the management of continental waters, its application
regional or national strategies at a local level.
to the coastal zone is more limited because it lacks a
The main aim of this article is to present a method for
corresponding geographical structure in the marine
classification of the coastal zone into regions by defining
domain. For example, in Catalonia the presence of
HEMUs. One of the characteristics of the approach is
only two large river basins means that this approach
that these units are integrated within the administrative
would not fully reflect the spatial variability of coastal
framework and can therefore be used as management
properties. Consequently, it has to be reduced to
units for implementing ICZM initiatives. The method is
smaller units to be a viable framework through which
applied to the Catalan coast of Spain to identify man-
to develop management plans (DMAH 2004). When
agement units in which specific planning strategies such
active administrative management units already exist
as the Coastal Zone Management Strategic Plan (PE-
in the coastal zone, an alternative approach is the
GICZ; DMAH 2004) and activities can be imple-
inclusion of the environmental values in the existing
mented according to the socioeconomic and natural
structure to provide an integrated model (Walpole
´ ´ characteristics of the territory.
1998, Barragan 2004, Sarda and others 2005). The need
for a detailed identification of management units on a
larger scale has led to the development of regional
Area of Study
initiatives based on detailed analysis and maps. Few
studies have used a combination of the two scales to
The Catalan coast is one of the richest and most rapidly
perform a HEMU regionalization, because of the
developing regions in Spain. Of the total population of
difficulty of integration in the ecosystem approach
´˜ Catalonia, 44% (2.79 million in 2001) lives in just 7%
(Yanez-Arancibia and Day 2004).
(70 municipalities) of the total surface area (IDESCAT
Indicator-based assessment and evaluation has
2005). The coastline is 699 km long and includes a wide
commonly been used to track the performance and
variety of temperate coastal systems. This results in
progress of ICZM plans and programs from a local to a
considerable geomorphological and biological diver-
national scale (Burbridge 1997, Belfiore 2003, Henoc-
sity. Figure 1 shows the administrative regions of the
que 2003, and references therein). Several methods
Catalan coastal area. Past and present human settle-
that incorporate multidimensional analysis have been
ments reflect the organization of socioeconomic activ-
used in the development of coastal classifications and
ities. The Mediterranean climate helped to configure
indices. As an example, Gornitz (1990) used a combi-
the current structure based on typical coastal activities
nation of methods ranging from geometric means to
such as tourism, commerce, agriculture, and more
factor analysis for classification of vulnerability and
recently, residential developments. Industrial and
generation of indices. For coastal indicators to be
commercial activities are strongly associated with the
effective in ICZM, it is necessary to demonstrate pro-
metropolitan areas of Barcelona (Central) and
gress and results in a comparable manner across spatial
Tarragona (South) but are less significant along the rest
scales and management levels (Belfiore 2003). Several
of the coast, where other economic activities (mainly
issues related to the scale problem have been identified
´
tourism) dominate (Sarda and others 2005).
in previous research, the modifiable aerial unit prob-
123
996 Environ Manage (2006) 38:993–1005
Fig. 1 Catalan coastal zone. Comarcas
and municipalities administrative
division
The Spanish coast is not only a complex area from a rational management of the coast. However, because of
physical, demographic, and economic point of view, but the diversity of the biophysical and socioeconomic
also because of the way it is regulated. There are three dimensions of the Catalan coast, it is difficult to
administrative levels in terms of institutions and legis- implement without a HEMU schema. Although the
lation: the central government of Spain, the regional importance of discrete planning units was stated in the
government of Catalonia, and the municipalities. objectives of the Catalan Agenda 21, the existing divi-
Within those levels, the Catalan coast is governed sion of legal and administrative responsibilities may
through two main legal instruments. First, the Spanish account for the lack of an effective HEMU framework.
National Coastal Law constitutes the jurisdictional There is a mismatch between the administrative
framework through which coastal zones are organized, units in the terrestrial and marine domains of the
specifically in terms of coastal public property (BOE coastal zone. In the terrestrial part there is a clear
1989). Despite the fact that this does not define man- spatial structure based on municipalities, whereas no
agement attributions to the Catalan coastal zone, it equivalent division exists in the marine domain. Fur-
does offer a general coastal zoning schema, as men- thermore, data with which to characterize the status of
tioned previously. The second instrument, the Statute the marine portion are scarce and heterogeneously
of the Autonomous Community of Catalonia, sets out distributed in comparison with a well-monitored ter-
the limited competencies of the Generalitat (regional restrial system. Moreover, most of the environmental
government) with respect to the Catalan coast and its status of the coastal zone is affected and/or controlled
marine environment (BOE 1979). Although in general by activities that take place in the terrestrial domain,
the Spanish government manages most activities re- such as urban development and tourism (Nunneri and
lated to the marine domain (as set out in the Coastal others 2005). Consequently, the scope of the present
Law), some of the activities (mainly seasonal services study is to identify inland territorial units with homo-
such as upkeep and cleaning of beaches) that influence geneous characteristics in which coastal managers have
the structure and dynamics of the shoreline (plus inte- responsibilities and in which they can develop a plan-
rior waters from base line) are managed by the local ning schema of priorities and implement strategies.
municipalities, which constitute the minimum admin- Specific typologies developed by scientific and
istrative and management implementation unit. Fol- management communities have been used in previous
lowing the EU recommendation on the implementation planning efforts. Such classifications are commonly
of integrated coastal zone management in Europe based on a single characteristic and have linear fea-
(COM/00/545), the Generalitat has already launched tures. The Master Ports Plan of the Generalitat is the
PEGIZC (DMAH 2004). This strategic plan constitutes most comprehensive coastal study undertaken in
a first step in a long-term move toward a much more Catalonia. It proposed a division of the coast into 21
123
Environ Manage (2006) 38:993–1005 997
data models for studies of coastal zones. However, GIS
continuous sectors based on homogeneous coastline
also face problems in effectively representing the coast
typologies, later classified into six geomorphological
(Mueller and others 2002), and data model and struc-
coastal types (DPTOP 1983). A more recent initiative
ture have been identified by Bartlett (2000) as the two
is the Oil Spill Prevention Plan, which assessed the
major concerns in the development of a coastal infor-
vulnerability of the previous 21 coastal sectors based
mation system.
on the composition of their benthic communities. The
Most existing studies of coastal area classification
criteria of the plan are (1) exposure to marine hydro-
use the shoreline as the basic representation unit. In
dynamics, (2) functional value per se for the ecosystem,
this shoreline-oriented approach, the explicit spatial
(3) rarity, and (4) ecological resilience (CAMCAT;
structure of system properties and dynamics is lost, and
DMAH 2003). Other landscape units have been iden-
only the resulting classification is retained. This is
tified through a region-specific analysis, e.g., the envi-
equivalent to assigning the entire properties of the
ronmental transformation of the northern Catalan
coastal area to a given length of shoreline without
coast or Costa Brava. Although units were defined
maintaining the original spatial reference (DPTOP
using an aggregation criterion of the geomorphology
1983, Fricker and Forbes 1988, Maxwell and Bud-
matrix based on current human perception of such
´ demeier 2002, DMAH 2003, Vafeidis and others 2004).
landscapes (Nogue 2004), classifications were restricted
However, linear-feature models are commonly used in
to one dimension (e.g., the natural environment) and
coastal mapping and analysis (Shupeng 1988), based on
lacked aspects of integration with socioeconomic
the common perception of the coast as a linear entity,
activities. In the neighboring French Mediterranean,
which assumes that its two horizontal dimensions are
the coast has been divided into 50 homogeneous zones
essentially equivalent (Goodchild 2000). This repre-
within the context of the Master Plan for the Devel-
sents one of the main limitations of the data model,
opment and Management of Water (SDAGE; RCM-
which fails to address problems of variable spatial
Comite de Bassin 1995, Henocque and Andral 2003).
resolution of coastal data (Vafeidis and others 2004).
Although the divisions are based on coastal geomor-
The aim of the present study is to develop a
phology, they have been used by the regional water
framework of geospatial coastal units that can be used
agency for more than 10 years to monitor water
in integrated management and that extends beyond the
quality.
shoreline level. Because of the spatial scale of the
relevant elements and the management model that will
be implemented in Catalonia, the management units
Methodological Approach
are based on a polygon data model in which discrete
The Geographic Information System units represent subsystems whose processes and func-
tions (including morphometric capabilities) can be
In order to develop a HEMU-based regionalization, subject to assessment, modeling, and monitoring
the terrestrial coastal subsystem was divided into nat- (Bartlett 2000). Few thematic mapping efforts have
ural (biophysical) and socioeconomic dimensions been undertaken in Catalonia. Although the descrip-
according to the generally accepted ICZM framework. tors were created from the available data (published
Because of the heterogeneity of this area and the need mainly by the local government), some of the spatial
to incorporate the environmental structure and func- representations were developed by the Coastal Man-
tion effectively, a regional, subnational cartographic agement Area of the LIM-UPC. To incorporate them
scale between 1:25,000 and 1:50,000 was chosen for the into the Catalan Coastal GIS (which began to be
developed in 2003 using ArcViewTM v3.x software
purpose of the study according to UNEP (1995)
recommendations. from ESRI), spatial data layers obeyed quality stan-
The complex nature of the coast presents a chal- dardization processes for format, scale, and metadata.
lenge for the determination of appropriate structures
for use when analytical and information frameworks Environmental Descriptors
are needed. This multidimensional spatial complexity
can be addressed more efficiently with the aid of geo- It was assumed that variations in the environmental
graphic information systems (GIS; Shupeng 1988, state (or health) of the coastal zone are controlled by
Bartlett 2000, among others). Because the representa- spatial and temporal variations in the characteristics
tion of a system’s elements is an important factor for and processes of the system. Such changes are the re-
the organization of databases, GIS have been widely sult of interactions between human and biophysical
used to integrate topological terrestrial and marine subsystems (UNESCO 1997, DMAH 2004, Vafeidis
123
998 Environ Manage (2006) 38:993–1005
Table 1 Theme by dimension used for the Catalan coastal zone HEMU definition
Dimension Theme Cartographic scale Year Descriptor (s)
Inhabitants count1
Socioeconomic Population size 50,000 2004
Mean anual rate1
Population growth 50,000 2001
Euros at market price1
Gross National Product 50,000 1996
Hotel beds by population1
Accommodation coefficient 50,000 2002
Urban area and Infrastructure1,2
Impervious surface 50,000 2003
Protected areas & wetlands surface3
Natural Natural protected area 25,000 2004
Areas surface3
Geomorphologic relevance 50,000 2002
Naturalness, diversity, and rarity4
Vegetation condition 25,000 2004
Environmental degradation3
Landscape transformation 50,000 2004
River flow and quality3,5
Running water condition 50,000 2003
Coastal geomorphology and dynamics6
Coastal geomorphology 50,000 1983
Source: (1) Catalan Statistics Institute (IDESCAT-GenCat); (2) Blanes Advance Studies Center (CEAB-CSIC); (3) Department of
Environment and Housing of the Catalan Government (DMAH-GenCat); (4) Plant Biology Department of the University of Bar-
celona (UB); (5) Water Catalan Agency (ACA-GenCat); (6) Department of Land Policy and Public Works (DPTOP-GenCat)
and others 2004). These interactions are considered environmental concerns in Catalonia (loss of biodi-
within the Catalan PEGIZC by focusing on five of the versity, fresh and marine water quality, and habitat
seven specific objectives: consolidation of undeveloped condition and transformation).
land, sustainable land use, land-derived marine pollu- The natural dimension themes were incorporated at
tion, erosion mitigation, and biodiversity conservation the municipality level. However, the natural protected
(DMAH 2004). Themes were chosen on the basis of area and the geomorphological relevance themes were
their independent capacity to represent the coastal incorporated at the landward 200-m fringe. This ap-
issues and were used to build up a data-driven classi- proach tried to capture the functional processes that
fication process (bottom-up). As in the case of indica- comprise the strip 200-m inland from the shoreline in
tors, a reduced number of variables is desirable for order to capture the coastal dynamics; this character-
prediction of the environmental state (Meentemeyer istic guards against overestimation of real conservation
and Box 1987). The idea is to reproduce most of the and the condition of coastal resources. The 200-m strip
system dynamics with a minimum number of descriptor constitutes the coastal conservation easement zone
criteria. Thus, selected themes represent the demo- indicated in the Spanish Coastal Law (BOE 1989). The
graphics, economy, geographic and biological diversity, natural geospatial features were incorporated into the
water resources, and coastal geomorphology of the GIS using the original minimal mapping unit (as pro-
Catalan coast. A total of 11 geospatial themes were vided by the source, e.g., raw polygons), be they
selected according to their conceptual, environment- polygons, lines, or points, and were later aggregated at
specific contributions as quantifiable phenomena of the the municipality level. Municipalities are the smallest
dynamic coastal subsystem and the quality of the official geographical management unit, and they con-
available data. The quality-control schema was based stitute the highest administrative implementation level,
on the following criteria: (1) 1:50,000 subnational car- and therefore, the most effective planning unit for
´
tographic scale or larger, (2) whether the source was ICZM (Sarda and others 2005). In contrast, the themes
official or not, and (3) data update criteria. Table 1 corresponding to the socioeconomic dimension were
shows the themes used and their descriptors, the spatial georeferenced to the comarca (a territorial unit com-
scale and the year the data were gathered. parable to a county), because this constitutes the
Within the socioeconomic dimension, the gross na- highest administrative level for which there is complete
tional product (GNP) was the most robust indicator, and official statistical data, and because comarcas are
because of its capacity to integrate several elements of recognized as a real and practical administrative ter-
economic development, even though it was the least ritorial unit in Catalonia, as well as in the rest of Spain,
up-to-date dataset. The tourist industry is considered thereby providing an accepted spatial framework.
the most significant environmental influence on the Comarcas are groups of municipalities (cluster), and
´
Catalan coast (Sarda and others 2005); thus, the they were selected because a large part of the socio-
accommodation coefficient was included as a relevant economic data available is only complete for 68.5% of
socioeconomic factor. The group of themes corre- municipalities (those with more than 5000 residents).
sponding to the environmental dimension coincided Themes were spatially combined using the GIS to
with the main institutional and governmental produce an ordinal pseudo-indicator of a specific
123
Environ Manage (2006) 38:993–1005 999
tion of integrated coastal zone management in
desirable condition of each theme. The resulting con-
Europe (EC 2002).
tinuous real number scale for each theme was numeri-
• They should constitute local administrative (man-
cally aggregated into an arbitrary four-way
agement) units.
classification, whether or not it was originally on an
• They should be based on real, natural, biophysical
ordinal scale. Gornitz (1990) and Gornitz and others
data.
(1994) used a similar approach to develop indices of
• They should integrate and reflect the principal
several coastal characteristics that were aggregated into
existing structure and functional processes of the
a vulnerability index using a linear model. The classifi-
coastal environment.
cation method used the Jenks optimization, which
• They should be derived from a combination of
identifies break points between classes by minimizing
independent characteristics that remain constant
the sum of the variance within each of the classes (Jenks
over time (wherever possible).
1967). This method identifies groupings and patterns
inherent in the data and produces a more objective
The natural and socioeconomic rationalizations
aggregate representation of spatial variability, thus
were aggregated to form the final HEMU map. The
providing a valuable tool with which to explore and
aggregation process obeyed certain algebraic combi-
represent data by minimizing its natural variation (Smith
nation rules. The final regional HEMU map was pro-
1986).
duced using four category units for the 12 comarcas of
the Catalan coast. An additional analytical phase de-
Data Aggregation Method fined spatial modeling rules to determine criteria for a
proposed natural coastal resources conservation
The natural themes considered in the analysis scenario.
(Table 1) were aggregated at the level of the comarca
to be spatially coherent and consistent with the socio-
economic data scale. An aggregation method based on
Results
a weighted average was used to represent the contri-
bution of the surface area of coastal municipalities to
The implementation of the Catalan ICZM strategic
the comarca level for the natural dimension themes
plan requires a territory-based spatial framework,
(see Gornitz 1990 for a discussion of data aggregation
which in this case is based on the definition of HEMUs.
methods). This met the requirement to establish a
Although the coastal system consists of several differ-
common spatial framework and prevented inferences
ent dimensions that determine its stability and health,
from higher to lower levels of analysis that are asso-
only two are used in this study: the socioeconomic and
ciated with the ecological fallacy (Alker 1969, Cao and
natural dimensions. It was assumed that the Catalan
Lam 1997). Comarcas constitute true physical man-
coastal zone could be defined for management pur-
agement units, because they are based on the common
poses in terms of these two dimensions, consisting of
historical, cultural, and administrative characteristics
five and six themes respectively that were incorporated
of their constituent municipalities. They are therefore
in the GIS at cartographic scales of 1:25,000 to 1:50,000
important in ICZM planning and monitoring of the
(Table 1).
´
Spanish coast (Barragan 2004).
Table 2 shows the values generated by classifying
Catalan coastal comarcas using the Jenks method for
Regionalization Process
each theme and dimension. This classification is based
on results given in terms of ordinal classes, where the
Theme typologies were used to develop a specific
maximum value (four) indicates the highest relevance
regionalization map for each dimension. The algebraic
of the characteristic and the minimum (one) indicates
sum of individual themes represented the contribution
the lowest relevance. Table 2 also shows the surface
of the individual natural and socioeconomic regional-
area (in hectares) of the comarcas and provinces to
ization of the Catalan coast. The thematic map of each
indicate the relative geographical contribution of the
dimension represents an independent view of the ter-
themes in the regionalization process.
ritory, and together they constitute the main input for
Figure 2 shows the results of the socioeconomic and
the integrated regionalization process. The following
the natural thematic rationalizations. There is a clear
criteria form the basis of the HEMU definitions:
relationship between the two: in general, higher values
• They should follow the principles proposed in the for the socioeconomic component are accompanied by
EU recommendation concerning the implementa- lower values for the natural component. This pattern
123
1000 Environ Manage (2006) 38:993–1005
Table 2 Theme classification values by comarca of the Catalan coastal zone
Socioeconomic Natural
Comarca Province Has A B C D E F G H I J K
`
Alt Emporda Girona 135,697 1 2 1 3 1 3 3 4 4 2 1
`
Baix Emporda Girona 70,016 1 2 1 3 2 2 2 3 3 2 1
Selva Girona 99,5374 1 3 1 4 1 2 1 3 3 2 1
Maresme Barcelona 40,049 2 3 2 2 2 1 2 3 3 2 3
`
Barcelones Barcelona 14,463 4 1 4 I 4 1 2 2 1 2 2
Baix Llobregat Barcelona 48,664 3 2 3 1 3 2 2 3 2 2 4
Garraf Barcelona 18,503 1 4 1 2 2 2 2 3 3 2 3
´
Baix Penedes Tarragona 29,618 1 4 1 2 2 1 1 3 2 1 3
`
Tarragones Tarragona 31,931 1 2 2 3 3 2 2 3 2 2 2
Baix Camp Tarragona 69,633 1 2 1 2 1 2 1 4 3 1 3
Baix Ebre Tarragona 100,212 1 2 1 1 1 3 2 4 4 3 4
`
Montsia Tarragona 73,741 1 2 1 1 3 3 3 4 3 4
Themes: (A) Population size; (B) Population growth; (C) Gross National Product; (D) Accommodation coefficient; (E) Impervious
surface; (F) Natural protected area; (G) Geomorphologic relevance; (H) Vegetation condition; (I) Landscape transformation;
(J) Running water condition; (K) Coastal geomorphology
Fig. 2 Socio-economic and natural
regionalizations of the Catalan coast
clearly reflects the central role of the metropolitan integrated measurement of the two dimensions. How-
`
areas of Barcelona (Barcelones) and Tarragona (Tar- ever, this method of aggregation can introduce inter-
`
ragones) in the socioeconomic development of Cata- pretation errors, because zones with very different
lonia. The least developed areas in socioeconomic characteristics can have similar numerical values. Thus,
` `
terms correspond to those with the highest environ- Barcelones and Alt Emporda have an equal total
mental values and are located in the northern (Alt wealth value which in the first case is due to a high
` `
Emporda) and southern (Montsia) ends of the region, socioeconomic value and in the second is due to a
where the most important protected natural coastal high natural one.
areas are located (Cap de Creus and the Ebre delta, To prevent this, we used an integrative model in
respectively; see Figure 1). which the natural component was combined with the
Once these two independent rationalizations were socioeconomic component, but in which they were
performed, they were combined to define the map of inverse scaled (i.e., an original value of four is substi-
the HEMUs. Figure 3 shows the HEMUs obtained by tuted by a value of one) and averaged. The resulting
applying a method designed to retain the attribute values were obtained from the algebraic mean of both
homogeneity of units after aggregation. By applying a the regionalization of the dimensions and re-aggre-
direct averaging of the two dimensions, the numerical gating them to their class type (i.e., values ranging from
values attached to each comarca in Figure 3 should be 2.000 to 2.999 indicate class 2). Reclassified values were
obtained. This value, which we will refer to as ‘‘total assigned to a nonordinal nominal four-class scheme to
wealth,’’ is obtained by averaging natural and socio- avoid misinterpretation of results. The final results
economic values, and it can be considered an indicate units (comarcas) with similar socioeconomic
123
Environ Manage (2006) 38:993–1005 1001
Fig. 3 Homogeneous Environmental Management Units of the Fig. 4 Conservation HEMU regionalization scenario of the
Catalan coast. Numbers indicate total socio-economic and Catalan coast
natural total richness by unit
for management purposes. As in previous cases, the
and natural properties but without showing any prior- maximum value for the criteria selected is four; in this
case, the highest environmental values. Management
ity indication.
The four-class comarca map obtained represents a plans for these units should be properly considered.
reliable management regionalization of the Catalan
coast, while being a data-based and user-oriented
product. Based on the spatial aggregation method Discussion and Conclusions
developed, it was possible to account for the functional
homogeneity of the coastal zone. The HEMUs classify The GIS provided an appropriate geospatial structure
the comarcas into highly natural areas (A), seminatural through which to develop an efficient classification of
areas (B), semiurban areas (C), and areas with high coastal management units (Shupeng 1988). As sug-
socioeconomic development (D). Geographically, each gested by Bartlett (2000), GIS also played a key role in
of these classes (units) should be managed under a database construction, theme modeling, and visualiza-
desired ‘‘vision’’ that fulfills the expectations of the tion of results. Although the selected polygon data
population living in the area and obeys the established model does not account for the dimensional problems
legal framework. implicit in the line representation of the coast (Vafeidis
Finally, the need to incorporate a stronger plan for and others 2004), we also found that there is no
the conservation of natural resources in current and straightforward system to define an aerial model that
future coastal zone management strategies has been efficiently manages the dynamics of the two coastal
stressed previously by several authors (Sherman and dimensions studied (Mueller and others 2002). How-
Alexander 1986, Van der Weide 1993, EC 2002, ever, in this study we used the mean-based aggregation
DMAH 2004). A management scenario involving model proposed by Gornitz (1990), because it has been
environmental conservation was defined to conserve demonstrated to be less sensitive to data errors, omis-
the natural role of the coast and provide a tool for sions, and misclassifications.
managers that could contribute to the target set for In order to use a method that is general enough to
2010 by the Convention on Biological Diversity (CBD be applicable to most coastal zones, themes describing
1999). The scenario was defined by applying an arbi- each component were selected on the basis that they
trary relative weighting of 80% to the natural dimen- were relevant, georeferenced, and could be either
sion values and 20% to the socioeconomic values. easily measured or obtained from existing official data.
Figure 4 shows the resulting map of HEMUs in terms Although it might be desirable to integrate data at a
of ordinal values. In this case, the map represents con- larger cartographic scale, positive results were ob-
servation priorities for the Catalan coast. The regions served in the spatial patterns obtained at the Catalan
are clearly similar to those obtained from the equally coast geographic scale. This is clearly the result of the
weighted averaging map, with the differences between multisource database appropriate integration at a
them arising from the existence of priority indications subnational cartographic scale (1:25,000–1:50,000), as
123
1002 Environ Manage (2006) 38:993–1005
The integrated description of themes selected for
recommended by UNEP (1995) (Table 1). The themes
the Catalan coastal zone can be considered represen-
are relevant to most developed and developing coastal
tative of developed coastal areas, where high values for
zones, and only a few were specific to the coast ana-
the socioeconomic components are frequently accom-
lyzed. This approach differs from data-intensive studies
panied by low values for the natural components
requiring a large number of descriptors for each theme
(Figure 2) This also seems to confirm a global tendency
that in many cases prevent its practical application (see
in coastal areas for socioeconomic activities to gener-
an example in Cendrero and Fischer 1997). An exam-
ate significant pressures on coastal systems, leading to
ple of an area-specific variable is the accommodation
an inherent reduction in or degradation of natural re-
coefficient (number of hotel beds per inhabitant),
sources. A similar pattern was found in comarcas with
which is only relevant to areas in which tourism is an
high values for natural resources (the northern and
important economic activity. This is clearly the case for
southernmost comarcas); although these are the least
the Catalan coast, where tourism accounted for about
developed in socioeconomic terms, they were the
10.8% of GNP in 2002 (DCTC 2002). If this analysis
greatest contributors to the geographic and biological
were to be performed for a coastal zone with different
diversity of the Catalan coast. If a river subbasin
major economic sectors, the corresponding represen-
schema existed for this area, the present results could
tative indicator would need to be properly selected to
be complemented in the future with similar approaches
reflect the most important socioeconomic component.
´
to those used by Escofet (2002) and Yanez-Arancibia
In this study, two parameters in the natural dimension
and Day (2004).
were calculated for the 200-m-wide fringe along the
Whenever the natural and socioeconomic dimen-
coast using the GIS, instead of using municipalities as
sions have to be integrated in order to characterize the
spatial units. This was done to accurately reflect coastal
properties of discrete planning units, results can be
environmental resources and not environmental re-
unclear or susceptible to misinterpretation by manag-
sources in coastal administrative units (municipalities)
ers. This is due to the inverse relationship between the
in a specifically adapted ecosystem approach (Rappa-
socioeconomic and natural values of developed coastal
port 1999). This width corresponds to the official con-
areas mentioned above. Thus, two units with different
servation easement zone based on the administrative
characteristics (one with high socioeconomic and low
regulations for the Spanish coast (Spanish Coastal Law,
natural values, and the other with the reverse situation)
BOE 1989) and must therefore be adjusted to the spe-
could give the same overall integrated value if they
cific regulations of the coast to be analyzed.
were directly combined. Although the value obtained
The natural data layers at the level of the munici-
in this way could be interpreted as a measure of the
pality were aggregated at a higher administrative
total wealth (considering both themes) of the territo-
level—the comarca—by considering values corre-
rial unit, it is clear that the two units could not be
sponding to the number of coastal municipalities in-
managed in the same way. This problem was overcome
cluded in it. Thus, the use of comarcas, made up of
by reclassifying one of the components before adding
municipalities with similar characteristics, leads to a
them together and prevented the socioeconomic data
degree of uniformity that is most likely to be due to the
interval ranking problems experienced by McLaughlin
common natural and socioeconomic environment that
and others (2002). The implicit result of this operation
is implicit inside the boundaries, reflected in a unifica-
should be equivalent to only considering one of the two
tion effect within the comarca. This final geographic
components, and it can only be used for coasts that
scale was found to be useful for reducing the high var-
display the inverse relationship between socioeco-
iability found at the level of municipalities, which would
nomic and natural values mentioned above.
have complicated the design of an effective ICZM
The bottom-up approach used here provided a data-
strategic plan for Catalonia (or probably anywhere
driven environmental regionalization of the coast that
else). This scale still retained the major sources of var-
could not have been obtained with a predetermined
iability along the coast, and because data were upscaled
planning structure (Harff and Davis 1990). Thus, the
and no aerial subdivisions were made, it did not show
results obtained are not intended to provide a priori
significant MAUP symptoms (Marceau 1999). Like-
management priorities, but rather to identify classes of
wise, no scale-dependent problems were addressed in
truly homogeneous units that managers can use for
the classification process because several themes were
future planning, policy implementations, and moni-
compiled from the beginning at the comarca level and
toring initiatives. This can be seen clearly in Figure 3
not aggregated at a different resolution. Similar results
by comparing the difference in HEMU class (four
concerning the use of comarcas as aggregation planning
´ classes) with the total wealth values obtained (almost
units in Spain can be found in Barragan (2004).
123
Environ Manage (2006) 38:993–1005 1003
constant throughout the entire territory). However,
HEMUs with the lowest total wealth values (La Selva,
`
Maresme, Baix Penedes, and Baix Camp) should be
identified as critical hot spots in the ICZM strategic
plan. Compared with the rest of the territory, these hot
spots do not seem to have a dominant value or re-
source. As suggested by Burbridge (1997), a special
plan would have to be designed to improve their situ-
ation and to reach the average value throughout the
territory.
Following the recommendations of the Sixth Envi-
ronmental Action Programme of the European Com-
munity (EC), the conservation of natural resources has
been defined as a central objective of the Catalonia
ICZM strategic plan to maintain and/or improve the
environmental quality of the system and its associated Fig. 5 Touristic regionalization of the Catalan coast (DCTC
2002)
human societies (DMAH 2004). The specific conser-
vation regionalization developed here (Figure 4) pro-
vides a spatial vision based on the natural quality of the fails to reflect the actual socioeconomic and natural
coastal zone and at the same time serves to identify variability and complexity of the coastal zone. This
priority conservation areas, a process that has been generates five regions (Figure 5) that, despite being
proposed as relevant to coastal management by EC currently managed and exploited as homogeneous
(2002). According to the pattern observed, the areas units, are composed of comarcas with dissimilar
with the highest environmental values are the north- socioeconomic and natural values (Figures 2 and 3).
ernmost and southernmost comarcas, and conse- The method proposed here to define a multidimen-
quently, under the present management scenario, those sional HEMU-based regionalization of the coastal
are the areas with the lowest priority. For the comarcas zone using GIS overcomes these problems and can be
with the lowest natural values, two different manage- used to define a more integrated management plan.
ment options could be selected: defining immediate However, the present proposal represents the result of
actions for the improvement of environmental values a data-driven analysis, and the process should be
(condition) or abandoning them and converting them complemented by a more social vision that considers
into sacrificed areas in terms of natural wealth. The final the goals and interests of managers, stakeholders, and
choice will depend on the level of transformation shown end users.
by these areas, as well as local institutional capacity. In In summary, the regionalization process performed
any case, to build a management-oriented scenario, the here for the Catalan coastal zone generated four dif-
selection of weights for the socioeconomic and natural ferent classes of HEMUs, for which socioeconomic and
components should be based on real policy objectives natural characteristics were combined in a GIS to give
as part of a more systemic view (Van der Weide 1993). an overall integrated value. The GIS proved to be an
Thus, this study only represents a proposal for manag- efficient tool for data management, analysis, and visu-
ers to consider in relation to such issues. alization in the overall process of defining coastal
Although based on the comarca administrative management units. This HEMU-based regionalization
units, the regionalization of the Catalonia coastal zone of the territory is a way to rationalize the definition of
based on HEMUs performed here does not correspond the Catalan ICZM strategic plan. This geospatial ap-
to any other existing comarca-based regionalization of proach could also be adapted and applied to other
the area. Most existing regionalizations are based on a coastal regions. Finally, the relevance of the process
single theme (typology) and consequently fail to cap- will ultimately depend on specific management goals
ture the integrated structure and functioning of the and objectives, and must be considered in the context
coastal system. As an example, the Catalan coastal of the need for a multicomponent spatial vision of the
tourism regionalization (DCTC 2002) is based on the coastal system. The proposed HEMU regionalization,
major economic driving force for the coastal zone, e.g., based on the comarca as the administrative/manage-
the tourist industry. In spite of the relative weight of ment unit, is expected to be an important tool for the
this factor in the socioeconomic structure, using it as future implementation of the recent ICZM strategic
the only regionalization parameter for the territory plan for Catalonia.
123
1004 Environ Manage (2006) 38:993–1005
Acknowledgments This work has been carried out within the Cendrero A., D. W. Fischer. 1997. A procedure for assessing the
framework of the Mevaplaya project (REN2003-09029-C03-01/ environmental quality of coastal areas for planning and
MAR), which is funded by the Spanish Ministry of Education management. J Coastal Res 13:732–744
and Science. The first author was supported by a doctoral grant Christian C. S. 1958. The concept of land units and land systems.
´
from the National Science and Technology Council of Mexico Proc Ninth Pacific Congress 20:74–81
(CONACyT) and the second author by the University Research DCTC. 2002. Catalonia tourism in numbers 2002. Tourism studies of
Promotion Award for Young Researchers of the Government of Catalonia. Commerce, Tourism and Consumer Department
Catalonia. The authors wish to thank the organizations and (DCTC), Autonomous Government of Catalonia, Barcelona,
institutions that supplied the data used in this work. Special Spain, 8 pp
´
thanks are due to Albert Ferre (Universitat de Barcelona), DMAH. 2003. Marine accidental pollution emergency special
´
Ramon Jordana (Departament de Agricultura, Ramaderia plan of Catalonia (CAMCAT). Environment and Housing
`
i Pesca, Generalitat de Catalunya), Marta Manzanera (Agencia Department (DMAH), Autonomous Government of Cata-
´
Catalana de l’Aigua), and Xavier Martı (Departament de Medi lonia, Barcelona, Spain (CD-ROM)
Ambient i Habitatge, Generalitat de Catalunya). We are also DMAH. 2004. Integrated coastal zone management strategic
`
grateful to Modest Fluvia (Universitat de Girona) for his help in plan of Catalonia (PEGICZ). Environmental Section.
defining the economic dimension. We thank Y. Henocque and Environment and Housing Department (DMAH), Auton-
two other anonymous reviewers for their comments and sug- omous Government of Catalonia, Barcelona, Spain, 66 pp
gestions on the original manuscript. DPTOP. 1983. Ecological and environmental study of coastal
sections. In ports and Transportation General Direction,
Territorial Policy and Public Works Department (DPTOP)
(ed.) Recreational Ports Plan Study. Volume 2. Autonomous
References
Government of Catalonia, Barcelona, Spain, pp 273–390
EC. 2000. Council Directive of the European Parliament and the
Alker R. 1969. A typology of ecological fallacies. In Dogan M, Council of 23 October 2000. Establishing a framework for
M. Rokkan (eds), Quantitative ecological analysis in the community action in the field of water policy. Official
social sciences. MIT Press, Cambridge, Massachusetts, pp 3 Journal of the European Communities (2000/60/EC), Eur-
Amir S. 1987. Classification of coastal resources: a Mediterra- opean Commission (EC), Brussels, Belgium, 72 pp
nean case study. Landscape Urban Planning 14:399–414 EC. 2002. Recommendation of the European Parliament and of
Baja S., D. M. Chapman, D. Dragovich. 2002. A conceptual the council of 30 May 2002 concerning the implementation
model for defining and assessing land management units of integrated coastal zone management in Europe. Official
using a fuzzy modelling approach in GIS environment. En- Journal of the European Communities (2002/413/EC),
vir Manage 29:647–661 European Commission (EC), Brussels, Belgium, 4 pp
´
Barragan J. M. 2004. The littoral areas of Spain. From geo- EEA. 2003. DMEER: Digital Map of European Ecological Re-
graphic analysis to integrated management. Ariel S.A., gions. The European Topic Centre on Nature Protection
Barcelona, Spain, 214 pp. and Biodiversity. European Environmental Agency [online:
Bartlett D. 2000. Working on the frontiers of science: applying http://dataservice.eea.eu.int/atlas/viewdata/viewpub.asp?id=7],
GIS to the coastal zone. In Wright D, D. Bartlett (eds), revised on 20 April 2005
Marine and coastal geographical information systems. Tay- ´
Escofet A. 2002. Alternativas para la regionalizacion del espacio
lor & Francis, London, United Kingdom, pp 11–24 ´
marino de Mexico. Working document prepared for the
Bartley J. A., J. W. Buddemeier, D. A. Bennett. 2001. Coastline Mapping Marine and Estuarine Ecosystems of North
complexity: a parameter for functional classification of ´ ´
America Project. Centro de Investigacion Cientıfica y de
coastal environments. J Sea Res 46:87–97 ´
Educacion Superior de Ensenada & Commission for Envi-
Belfiore S. 2003. The growth of integrated coastal management ronmental Cooperation, NAFTA, Ensenada, Mexico, 13 pp
and the role of indicators in integrated coastal management: Finkl C. W. 2004. Coastal classification: systematic approaches to
introduction to the special issue. Ocean Coastal Manage consider in the development of a comprehensive scheme.
46:225–234 J Coast Res 20:166–213
Bian L. 1997. Multiscale nature of spatial data in scaling up Fricker A., D. L. Forbes. 1988. A system for coastal description
environmental models. In Quattrochi D. A, M. F. Goodchild and classification. Coast Manage 16:111–137
(eds), Scale in remote sensing and GIS. Lewis Publishers, Goodchild M. F. 2000. Foreword. In Wright D, D. Bartlett (eds),
Boca Raton, Florida, pp 13–26 Marine and coastal Geographical Information Systems.
BOE. 1979. Organic Law 4/1979, of December 18, Catalonia Taylor & Francis, London, United Kingdom, pp xiii–xv
Autonomous Statute. State Official Newsletter (BOE), Gornitz V. 1990. Vulnerability of the East Coast, U.S.A. to fu-
Published 22 December 1979, Madrid, Spain ture sea level rise. J Coast Res special issue no. 9:201–237
BOE. 1989. Law 22/1988, of July 29 (Leadership of the State) of Gornitz V. M., R. C. Daniels, T. W. Whites, K. R. Birdwell. 1994.
Coasts. State Official Newsletter (BOE), Legal Documents The development of a coastal risk assessment database:
Collection. Published January 1994, Madrid, Spain, 389 pp vulnerability to sea-level rise in the U.S. Southeast. J Coast
Burbridge P. R. 1997. A generic framework for measuring suc- Res special issue no. 12 (Coastal Hazards):327–338
cess in integrated coastal management. Ocean Coastal Harff J., J. C. Davis. 1990. Regionalization in geology by mul-
Manage 37:175–189 tivariate classification. Math Geol 22:573–588
Cao C., Lam N. 1997. Understanding the scale and resolution Henocque Y. 2003. Development of progress indicators for
effects in remote sensing and GIS. In Quattrochi D. A., M. coastal zone management in France. Ocean Coast Manage
F. Goodchild (eds), Scale in remote sensing and GIS. Lewis 46:363–379
Publishers, Boca Raton, Florida, pp 57–72 Henocque Y., B. Andral. 2003. The French approach to man-
CBD. 1999. Liaison group on the ecosystem approach. Con- aging water resources in the Mediterranean and the new
vention on Biological Diversity & UNESCO, Workshop European Water Framework Directive. Marine Pollut Bull
Report, September 15–17, 1999, Paris, 11 pp 47:155–161
123
Environ Manage (2006) 38:993–1005 1005
IDESCAT. 2005. 2005 Book of statistics. Statistics Institute of SDAGE, 295 pp. Volume 3: Cartography objectives and
Catalonia [online: http://www.idescat.net/], revised 22 April priorities, 15 A3 maps. Rhone-Mediterranean Watershed
2005 Delegation (RMC-Comite de Bassin), France
´
Jenks G. F. 1967. The data model concept in statistical mapping. Sarda R., C. Avila, J. Mora. 2005. A methodological approach to
Int Yearbook Cartogr 7:186–190 be used in integrated coastal zone management process: the
Marceau D. J. 1999. The scale issue in social and natural sciences. case of the Catalan Coast (Catalonia, Spain). Estuarine
Can J Remote Sens 25:347–356 Coastal Shelf Sci 62:427–439
Maxwell B. A., R. W. Buddemeier. 2002. Coastal typology Sherman K., L. M. Alexander. 1986. Variability and manage-
development with heterogeneous data sets. Regional Envir ment of large marine ecosystems. AAAS selected sympo-
Change 3:77–87 sium 99. Westview Press, Boulder, Colorado, 319 pp
McLaughlin S., J. McKenna, J. A. G. Cooper. 2002. Socio-eco- Shupeng C. 1988. The coastline as a base for global databases: a
nomic data in coastal vulnerability indices: constrains and pilot study in China. In Tomlinson R (ed), Building data-
opportunities. J Coast Res 36:487–497 bases for global science. Taylor & Francis, London, United
Mee L. D. 2005. Asssessment and monitoring requirements for Kingdom, pp 202–215
the adaptive management of Europe’s Regional Seas. In Smith R. M. 1986. Comparing traditional methods for selecting
Salomons W, Vermaat J, K. Turner (eds), Managing Euro- class intervals on choropleth maps. Professional Geographer
pean coasts: past, present and future. Environmental Sci- 38:62–67
ences Series, Springer-Verlag, Berlin, Germany, pp 227–237 UNEP. 1995. The development and implementation of ICAM.
Meentemeyer V., E. O. Box. 1987. Scale effects in landscape In UNEP regional seas reports and studies (ed), Guidelines
studies. In M. G. Turner (ed), Landscape heterogeneity and for integrated management of coastal and marine areas:
disturbance. Springer-Verlag, New York, New York, pp 16–34 with special reference to the Mediterranean basin. PAP/
Mueller M., B. Meissner, W. Weinrebe. 2002. TerraMarIS–Ter- RAC (MAP-UNEP), no. 161, Split, Croatia, pp 19–33
restrial and Marine Information System. In Breman J (ed), UNESCO. 1997. Definition of the coherent management units:
Marine geography: GIS for the oceans and the seas. ESRI Stage 2. In Methodological guide to integrated coastal zone
Press, Redlands, California, pp 92–102 management. Manuals & guides 36. Intergovernmental
´ ´
Nogue J. 2004. The territorial and landscape transformation of Oceanographic Comision, France, pp 16–19
the Costa Brava (1956-2003): Present situation and pro- Vafeidis A. T., R. J. Nicholls, L. McFadden, J. Hinkel, P. S.
posed activities. In workshop 1: Scarce Territory or Fragile Grashoff. 2004. Developing a global database for coastal
landscape. Costa Brava Debate (March 5, 2004) [online: vulnerability analysis: design issues and challenges. In XX
http://www.debatcostabrava.org/], revised on 15 November ISPRS Congress (ed), The international archives of the
2005. Rosas, Spain photogrammetry, remote sensing and spatial information
Nunneri C., K. R. Turner, A. Cieslak, A. Kannen, R. Klein, L. sciences, vol. 34, part XXXV, commission IV, 12–23 July,
Ledoux, J. Marquenie, L. Mee, S. Moncheva, R. Nicholls, Istanbul, Turkey, pp 801–805
´
W. Salomons, R. Sarda, M. Stive, T. Vellinga. 2005. Inte- Van der Weide J. 1993. A systems view of integrated coastal
grated assessment and future scenarios for the coast. In management. Ocean Coast Manage 21:129–148
Vermaat J, L. Bouwer, K. Turner, W. Salomons (eds), Walpole S. C. 1998. Integration of economic and biophysical
Managing European coasts: past, present and future. Envi- information to assess the site-specific profitability of land
ronmental Sciences Series, Springer-Verlag, Berlin, Ger- management programmes using a GIS. In Proceedings of
many, pp 271–290 the Eighth ISCO Conference, New Delhi, India, pp 1663–
Openshaw S. 1984. The modifiable aerial unit problem. CAT- 1669
´˜
MOG 38. Geobooks, Norwich, United Kingdom Yanez-Arancibia A., J. W. Day. 2004. Environmental sub-
Rappaport J. 1999. The ecosystem approach from a practical regions in the Gulf of Mexico coastal zone: the ecosystem
point of view. Conserv Biol 13:679–681 approach as an integrated management tool. Ocean Coast
RMC-Comite de Bassin. 1995. Zoning and management scheme Manage 47:727–757
of the Rhone-Mediterranean and Corsica basin (SDAGE). Zonneveld I. S. 1994. Basic principles of classification. In Klijn F
Users’ guide, 13 pp. Volume 1: Key fundamental orienta- (ed), Ecosystem classification for environmental manage-
tions, operational measures and modalities, 120, pp: Volume ment. Kluwer Academic Publishers, The Netherlands,
2: Thematic archives, rules and recommendations of pp 23–47
123
DOI 10.1007/s00267-005-0210-6
Definition of Homogeneous Environmental Management Units
for the Catalan Coast
Jorge Brenner Æ Jose A. Jimenez Æ Rafael Sarda
´ ´
Received: 12 July 2005 / Accepted: 29 May 2006
Ó Springer Science+Business Media, Inc. 2006
Abstract Geographical areas constitute the basic Keywords Environmental regionalization Æ
implementation locus for integrated coastal zone man- Homogeneous Environmental Management Units
agement strategies and activities. Because the definition (HEMUs) Æ Integrated Coastal Zone Management
of territorial planning objectives may be affected by (ICZM) Æ Geographic Information System (GIS)
socioeconomic and environmental characteristics, one
of the main steps in the process involves dividing the
coast into homogeneous environmental management Introduction
units (HEMUs). This article presents a general and
simple method for regionalizing the landside of a Management of coastal areas under the sustainable
coastal zone into HEMUs and illustrates it through regional development mandate is a complex process.
application to the Catalan coast. Socioeconomic and Difficulties arise from the need to strike a balance
natural (biophysical) subsystems were selected as the between socioeconomic development and coastal con-
most appropriate dimensions of the regionalization servation. This balance may vary due to the high var-
process. Dimensions were described using 11 spatial iability of the primary components of the coastal
themes, which were managed in a geographic informa- system, i.e., the natural and socioeconomic subsystems
tion system environment that proved to be an adequate (Van der Weide 1993). The aim of integrated coastal
tool for the purpose. A final coastal zone map of four management is to maintain a sustainable relationship
classes of HEMUs connected to local administrative between the resources of these two subsystems and
units was obtained, and because it reflects the current their exploitation, preventing (or mitigating) potential
natural and socioeconomic dynamics, it can be consid- conflicts and reducing the uncertainties associated with
ered as an initial step in the planning process for the planning and decision making. However, to manage a
Catalan coast. Although the proposed method was coastal region properly, a clear picture should already
developed based on the characteristics of the Catalan have been obtained of the expectations of stakeholders
coast, it is general enough to be adapted and applied to and/or society regarding each specific unit of territory,
most developed or developing coastal areas. as well as the legal framework into which it fits and the
existing property rights (Mee 2005). When this vision is
shared and accepted, specific criteria can be developed
J. Brenner (&) Æ J. A. Jimenez to accommodate uses of coastal areas, to resolve po-
´
Laboratori d’Enginyeria Marıtima, ETSECCPB,
tential conflicts, and to facilitate the decision-making
`
Universitat Politecnica de Catalunya,
process. In Spain, the coastal zone is administratively
`
C/ Jordi Girona, 1-3, Modul D-1, 08034, Barcelona, Spain
defined in the Coastal Law (BOE 1989) in terms of a
E-mail: jorge.brenner@upc.edu
marine and terrestrial zone that falls within the public
´
R. Sarda
domain. It is a very narrow fringe of territory delimited
Centre d’Estudis Avancats de Blanes, CSIC
¸
on the land side by the innermost high-water level.
´
Carrer d’Acces a la Cala Sant Francesc, 14, 17300
Inland there is a conservation easement fringe of
Blanes, Spain
123
994 Environ Manage (2006) 38:993–1005
human activity. Several analytical approaches have
variable width with different restrictions. Although this
been used, such as multivariate classifications/cluster-
implies some kind of management or regulation of
ing, factor analysis, fuzzy logic, multicriteria analysis,
activities, there is an overlap with the responsibilities
and spatial overlapping (Fricker and Forbes 1988,
of the regional and local administrations. These factors
Gornitz 1990, Bartley and others 2001, Baja and others
generate a relatively poorly defined area in terms of
2002, Escofet 2002, Maxwell and Buddemeier 2002,
planning and management.
Henocque and Andral 2003, Vafeidis and others 2004,
Integrated coastal zone management (ICZM) is a
´˜
Yanez-Arancibia and Day 2004). The most common
tool to help achieve sustainable regional development
HEMU definitions have been based on biophysical
in coastal areas. The main purpose of all ICZM ini-
characteristics such as geomorphology, climate, vege-
tiatives is to maintain, restore, or improve specific as-
tation, and biodiversity. However, in order to develop
pects of coastal zone systems and their associated
an integrated vision of the coastal zone, the socioeco-
human societies. An important feature of ICZM ini-
nomic dimension needs to be incorporated into the
tiatives is that they address the needs of both socio-
´
process (Sarda and others 2005).
economic development and natural conservation in
As a starting point, typologies constitute repeatable
geographically specific planning activities at multiple
homogeneous units that are the basis for division or
administrative levels. Thus, geographic areas constitute
classification into geographical units. Usually, devel-
the basic implementation locus of ICZM strategies and
opment of a typology for geospatial data takes either a
activities. Many authors have emphasized the role of
top-down or a bottom-up approach (Maxwell and
appropriate territorial information and organized,
Buddemeier 2002). The top-down approach to classi-
coherent databases as essential for decision making in
fication is based on a decision tree containing prede-
the coastal zone (e.g., Shupeng 1988, Bartlett 2000).
fined environmental characteristics that is specifically
The coastal zone is characterized by a high degree of
developed for a given environment (e.g., Finkl 2004).
natural and socioeconomic heterogeneity because of
In the bottom-up approach, a clustering method is used
the existence of multiple resources and uses, and its
to identify groups with similar environmental charac-
highly dynamic nature (McLaughlin and others 2002).
teristics. A variation on bottom-up classification is the
The spatial heterogeneity of the coastal zone can be
regionalization approach, which locates spatially con-
rationalized by selecting homogeneous environmental
tiguous class members after clustering without atten-
management units (HEMUs), discrete homogeneous
tion to spatial location (Harff and Davis 1990).
areas, or units with similar characteristics (for a
Regions, which constitute a unique discrete system,
description of similar approaches, see Christian 1958,
become planning units and can be identified by a spe-
Amir 1987, UNESCO 1997). These territorial units
cific valuable quantifiable phenomenon. A combina-
should then be linked to a strategic territorial plan, and
tion of structural and functional typologies can
thus, to active management units (Mee 2005). These
determine the specific processes that constitute indi-
units form the basis for research and data collection,
vidual management regions. A review of the biophys-
and subsequently become the boundaries defining
ical characteristics used for the classification of coastal
areas with similar land attributes selected as decision
and marine environments can be found in Finkl (2004).
criteria for planning and evaluation (Baja and others
The analytical process leading to regionalization can be
2002). This process of reducing spatial complexity is a
divided into two discretization strategies: hierarchical
way of linking management decisions to the biophysi-
´˜
unit grouping and segmenting (Yanez-Arancibia and
cal and socioeconomic properties of a territory, and
Day 2004). These two approaches tend to give rise to
thus, meets the need of policy makers to access quan-
regions based on a hierarchical criterion of belonging
titative information on physical areas. To be an effi-
to a higher scale unit; thus, units can be identified as
cient management tool, these should also be integrated
either belonging to a higher region or forming one
within the existing administrative framework. To
(Escofet 2002). The interactions between the individ-
properly define a HEMU, natural and socioeconomic
ual regions should determine the territorial planning
properties must represent the coastal system as closely
schema that management needs for the process of
as possible (Zonneveld 1994), and if they are imple-
reconciling the natural and socioeconomic subsystems.
mented in a geospatial management framework, all the
Because of the difficulties associated with this data-
elements of the system (natural, socioeconomic,
driven process, most planning instruments, such as
administrative, etc.) must be spatially coherent.
assessment and evaluation, lack this framework.
The definition of HEMUs is a common task
As a useful working concept, the ecosystem
when one is dealing with systems with different
approach has been defined to help in the process of
environmental properties that support significant
123
Environ Manage (2006) 38:993–1005 995
lem (MAUP) being one of the most notable. MAUP
setting environmental management boundaries (CBD
appears in spatially averaged studies when units are
1999). Large-scale applications of this approach can be
subdivided into smaller nonoverlapping units such that
found in the different global proposals for environ-
intrinsic geographical meaning is absent (Openshaw
mental regionalization, such as the large marine eco-
1984). It has major implications in two areas: (1) the
systems of the world (Sherman and Alexander 1986)
number of aerial divisions of a unit that can be per-
and the environmental land units of the European
formed, and (2) the data aggregation at different
ecological regions proposals (EEA 2003). This ap-
resolutions (Bian 1997, Cao and Lam 1997). Although
proach recognizes the dynamics and complexity of
several solutions have been proposed, the main
ecosystems in order to provide an analytical frame-
uncertainty arises when geospatial data are scale
work for the development of managerial strategies
dependent (Cao and Lam 1997, Marceau 1999). Its
(Rappaport 1999). An example of regionalization is the
importance increases with increasing spatial and tem-
use of river basins to define management units for use
´˜ poral heterogeneity of the coast, and the difficulty of
in a variety of approaches (e.g., Yanez-Arancibia and
combining natural and socioeconomic subsystem indi-
Day 2004). This approach is used in the European
cators in the assessment process further complicates
Union (EU) to apply the European Directive on Water
the final situation. Consequently, methodological dif-
Policy (EC 2000). Although this approach is logical for
ficulties are presented for the implementation of
the management of continental waters, its application
regional or national strategies at a local level.
to the coastal zone is more limited because it lacks a
The main aim of this article is to present a method for
corresponding geographical structure in the marine
classification of the coastal zone into regions by defining
domain. For example, in Catalonia the presence of
HEMUs. One of the characteristics of the approach is
only two large river basins means that this approach
that these units are integrated within the administrative
would not fully reflect the spatial variability of coastal
framework and can therefore be used as management
properties. Consequently, it has to be reduced to
units for implementing ICZM initiatives. The method is
smaller units to be a viable framework through which
applied to the Catalan coast of Spain to identify man-
to develop management plans (DMAH 2004). When
agement units in which specific planning strategies such
active administrative management units already exist
as the Coastal Zone Management Strategic Plan (PE-
in the coastal zone, an alternative approach is the
GICZ; DMAH 2004) and activities can be imple-
inclusion of the environmental values in the existing
mented according to the socioeconomic and natural
structure to provide an integrated model (Walpole
´ ´ characteristics of the territory.
1998, Barragan 2004, Sarda and others 2005). The need
for a detailed identification of management units on a
larger scale has led to the development of regional
Area of Study
initiatives based on detailed analysis and maps. Few
studies have used a combination of the two scales to
The Catalan coast is one of the richest and most rapidly
perform a HEMU regionalization, because of the
developing regions in Spain. Of the total population of
difficulty of integration in the ecosystem approach
´˜ Catalonia, 44% (2.79 million in 2001) lives in just 7%
(Yanez-Arancibia and Day 2004).
(70 municipalities) of the total surface area (IDESCAT
Indicator-based assessment and evaluation has
2005). The coastline is 699 km long and includes a wide
commonly been used to track the performance and
variety of temperate coastal systems. This results in
progress of ICZM plans and programs from a local to a
considerable geomorphological and biological diver-
national scale (Burbridge 1997, Belfiore 2003, Henoc-
sity. Figure 1 shows the administrative regions of the
que 2003, and references therein). Several methods
Catalan coastal area. Past and present human settle-
that incorporate multidimensional analysis have been
ments reflect the organization of socioeconomic activ-
used in the development of coastal classifications and
ities. The Mediterranean climate helped to configure
indices. As an example, Gornitz (1990) used a combi-
the current structure based on typical coastal activities
nation of methods ranging from geometric means to
such as tourism, commerce, agriculture, and more
factor analysis for classification of vulnerability and
recently, residential developments. Industrial and
generation of indices. For coastal indicators to be
commercial activities are strongly associated with the
effective in ICZM, it is necessary to demonstrate pro-
metropolitan areas of Barcelona (Central) and
gress and results in a comparable manner across spatial
Tarragona (South) but are less significant along the rest
scales and management levels (Belfiore 2003). Several
of the coast, where other economic activities (mainly
issues related to the scale problem have been identified
´
tourism) dominate (Sarda and others 2005).
in previous research, the modifiable aerial unit prob-
123
996 Environ Manage (2006) 38:993–1005
Fig. 1 Catalan coastal zone. Comarcas
and municipalities administrative
division
The Spanish coast is not only a complex area from a rational management of the coast. However, because of
physical, demographic, and economic point of view, but the diversity of the biophysical and socioeconomic
also because of the way it is regulated. There are three dimensions of the Catalan coast, it is difficult to
administrative levels in terms of institutions and legis- implement without a HEMU schema. Although the
lation: the central government of Spain, the regional importance of discrete planning units was stated in the
government of Catalonia, and the municipalities. objectives of the Catalan Agenda 21, the existing divi-
Within those levels, the Catalan coast is governed sion of legal and administrative responsibilities may
through two main legal instruments. First, the Spanish account for the lack of an effective HEMU framework.
National Coastal Law constitutes the jurisdictional There is a mismatch between the administrative
framework through which coastal zones are organized, units in the terrestrial and marine domains of the
specifically in terms of coastal public property (BOE coastal zone. In the terrestrial part there is a clear
1989). Despite the fact that this does not define man- spatial structure based on municipalities, whereas no
agement attributions to the Catalan coastal zone, it equivalent division exists in the marine domain. Fur-
does offer a general coastal zoning schema, as men- thermore, data with which to characterize the status of
tioned previously. The second instrument, the Statute the marine portion are scarce and heterogeneously
of the Autonomous Community of Catalonia, sets out distributed in comparison with a well-monitored ter-
the limited competencies of the Generalitat (regional restrial system. Moreover, most of the environmental
government) with respect to the Catalan coast and its status of the coastal zone is affected and/or controlled
marine environment (BOE 1979). Although in general by activities that take place in the terrestrial domain,
the Spanish government manages most activities re- such as urban development and tourism (Nunneri and
lated to the marine domain (as set out in the Coastal others 2005). Consequently, the scope of the present
Law), some of the activities (mainly seasonal services study is to identify inland territorial units with homo-
such as upkeep and cleaning of beaches) that influence geneous characteristics in which coastal managers have
the structure and dynamics of the shoreline (plus inte- responsibilities and in which they can develop a plan-
rior waters from base line) are managed by the local ning schema of priorities and implement strategies.
municipalities, which constitute the minimum admin- Specific typologies developed by scientific and
istrative and management implementation unit. Fol- management communities have been used in previous
lowing the EU recommendation on the implementation planning efforts. Such classifications are commonly
of integrated coastal zone management in Europe based on a single characteristic and have linear fea-
(COM/00/545), the Generalitat has already launched tures. The Master Ports Plan of the Generalitat is the
PEGIZC (DMAH 2004). This strategic plan constitutes most comprehensive coastal study undertaken in
a first step in a long-term move toward a much more Catalonia. It proposed a division of the coast into 21
123
Environ Manage (2006) 38:993–1005 997
data models for studies of coastal zones. However, GIS
continuous sectors based on homogeneous coastline
also face problems in effectively representing the coast
typologies, later classified into six geomorphological
(Mueller and others 2002), and data model and struc-
coastal types (DPTOP 1983). A more recent initiative
ture have been identified by Bartlett (2000) as the two
is the Oil Spill Prevention Plan, which assessed the
major concerns in the development of a coastal infor-
vulnerability of the previous 21 coastal sectors based
mation system.
on the composition of their benthic communities. The
Most existing studies of coastal area classification
criteria of the plan are (1) exposure to marine hydro-
use the shoreline as the basic representation unit. In
dynamics, (2) functional value per se for the ecosystem,
this shoreline-oriented approach, the explicit spatial
(3) rarity, and (4) ecological resilience (CAMCAT;
structure of system properties and dynamics is lost, and
DMAH 2003). Other landscape units have been iden-
only the resulting classification is retained. This is
tified through a region-specific analysis, e.g., the envi-
equivalent to assigning the entire properties of the
ronmental transformation of the northern Catalan
coastal area to a given length of shoreline without
coast or Costa Brava. Although units were defined
maintaining the original spatial reference (DPTOP
using an aggregation criterion of the geomorphology
1983, Fricker and Forbes 1988, Maxwell and Bud-
matrix based on current human perception of such
´ demeier 2002, DMAH 2003, Vafeidis and others 2004).
landscapes (Nogue 2004), classifications were restricted
However, linear-feature models are commonly used in
to one dimension (e.g., the natural environment) and
coastal mapping and analysis (Shupeng 1988), based on
lacked aspects of integration with socioeconomic
the common perception of the coast as a linear entity,
activities. In the neighboring French Mediterranean,
which assumes that its two horizontal dimensions are
the coast has been divided into 50 homogeneous zones
essentially equivalent (Goodchild 2000). This repre-
within the context of the Master Plan for the Devel-
sents one of the main limitations of the data model,
opment and Management of Water (SDAGE; RCM-
which fails to address problems of variable spatial
Comite de Bassin 1995, Henocque and Andral 2003).
resolution of coastal data (Vafeidis and others 2004).
Although the divisions are based on coastal geomor-
The aim of the present study is to develop a
phology, they have been used by the regional water
framework of geospatial coastal units that can be used
agency for more than 10 years to monitor water
in integrated management and that extends beyond the
quality.
shoreline level. Because of the spatial scale of the
relevant elements and the management model that will
be implemented in Catalonia, the management units
Methodological Approach
are based on a polygon data model in which discrete
The Geographic Information System units represent subsystems whose processes and func-
tions (including morphometric capabilities) can be
In order to develop a HEMU-based regionalization, subject to assessment, modeling, and monitoring
the terrestrial coastal subsystem was divided into nat- (Bartlett 2000). Few thematic mapping efforts have
ural (biophysical) and socioeconomic dimensions been undertaken in Catalonia. Although the descrip-
according to the generally accepted ICZM framework. tors were created from the available data (published
Because of the heterogeneity of this area and the need mainly by the local government), some of the spatial
to incorporate the environmental structure and func- representations were developed by the Coastal Man-
tion effectively, a regional, subnational cartographic agement Area of the LIM-UPC. To incorporate them
scale between 1:25,000 and 1:50,000 was chosen for the into the Catalan Coastal GIS (which began to be
developed in 2003 using ArcViewTM v3.x software
purpose of the study according to UNEP (1995)
recommendations. from ESRI), spatial data layers obeyed quality stan-
The complex nature of the coast presents a chal- dardization processes for format, scale, and metadata.
lenge for the determination of appropriate structures
for use when analytical and information frameworks Environmental Descriptors
are needed. This multidimensional spatial complexity
can be addressed more efficiently with the aid of geo- It was assumed that variations in the environmental
graphic information systems (GIS; Shupeng 1988, state (or health) of the coastal zone are controlled by
Bartlett 2000, among others). Because the representa- spatial and temporal variations in the characteristics
tion of a system’s elements is an important factor for and processes of the system. Such changes are the re-
the organization of databases, GIS have been widely sult of interactions between human and biophysical
used to integrate topological terrestrial and marine subsystems (UNESCO 1997, DMAH 2004, Vafeidis
123
998 Environ Manage (2006) 38:993–1005
Table 1 Theme by dimension used for the Catalan coastal zone HEMU definition
Dimension Theme Cartographic scale Year Descriptor (s)
Inhabitants count1
Socioeconomic Population size 50,000 2004
Mean anual rate1
Population growth 50,000 2001
Euros at market price1
Gross National Product 50,000 1996
Hotel beds by population1
Accommodation coefficient 50,000 2002
Urban area and Infrastructure1,2
Impervious surface 50,000 2003
Protected areas & wetlands surface3
Natural Natural protected area 25,000 2004
Areas surface3
Geomorphologic relevance 50,000 2002
Naturalness, diversity, and rarity4
Vegetation condition 25,000 2004
Environmental degradation3
Landscape transformation 50,000 2004
River flow and quality3,5
Running water condition 50,000 2003
Coastal geomorphology and dynamics6
Coastal geomorphology 50,000 1983
Source: (1) Catalan Statistics Institute (IDESCAT-GenCat); (2) Blanes Advance Studies Center (CEAB-CSIC); (3) Department of
Environment and Housing of the Catalan Government (DMAH-GenCat); (4) Plant Biology Department of the University of Bar-
celona (UB); (5) Water Catalan Agency (ACA-GenCat); (6) Department of Land Policy and Public Works (DPTOP-GenCat)
and others 2004). These interactions are considered environmental concerns in Catalonia (loss of biodi-
within the Catalan PEGIZC by focusing on five of the versity, fresh and marine water quality, and habitat
seven specific objectives: consolidation of undeveloped condition and transformation).
land, sustainable land use, land-derived marine pollu- The natural dimension themes were incorporated at
tion, erosion mitigation, and biodiversity conservation the municipality level. However, the natural protected
(DMAH 2004). Themes were chosen on the basis of area and the geomorphological relevance themes were
their independent capacity to represent the coastal incorporated at the landward 200-m fringe. This ap-
issues and were used to build up a data-driven classi- proach tried to capture the functional processes that
fication process (bottom-up). As in the case of indica- comprise the strip 200-m inland from the shoreline in
tors, a reduced number of variables is desirable for order to capture the coastal dynamics; this character-
prediction of the environmental state (Meentemeyer istic guards against overestimation of real conservation
and Box 1987). The idea is to reproduce most of the and the condition of coastal resources. The 200-m strip
system dynamics with a minimum number of descriptor constitutes the coastal conservation easement zone
criteria. Thus, selected themes represent the demo- indicated in the Spanish Coastal Law (BOE 1989). The
graphics, economy, geographic and biological diversity, natural geospatial features were incorporated into the
water resources, and coastal geomorphology of the GIS using the original minimal mapping unit (as pro-
Catalan coast. A total of 11 geospatial themes were vided by the source, e.g., raw polygons), be they
selected according to their conceptual, environment- polygons, lines, or points, and were later aggregated at
specific contributions as quantifiable phenomena of the the municipality level. Municipalities are the smallest
dynamic coastal subsystem and the quality of the official geographical management unit, and they con-
available data. The quality-control schema was based stitute the highest administrative implementation level,
on the following criteria: (1) 1:50,000 subnational car- and therefore, the most effective planning unit for
´
tographic scale or larger, (2) whether the source was ICZM (Sarda and others 2005). In contrast, the themes
official or not, and (3) data update criteria. Table 1 corresponding to the socioeconomic dimension were
shows the themes used and their descriptors, the spatial georeferenced to the comarca (a territorial unit com-
scale and the year the data were gathered. parable to a county), because this constitutes the
Within the socioeconomic dimension, the gross na- highest administrative level for which there is complete
tional product (GNP) was the most robust indicator, and official statistical data, and because comarcas are
because of its capacity to integrate several elements of recognized as a real and practical administrative ter-
economic development, even though it was the least ritorial unit in Catalonia, as well as in the rest of Spain,
up-to-date dataset. The tourist industry is considered thereby providing an accepted spatial framework.
the most significant environmental influence on the Comarcas are groups of municipalities (cluster), and
´
Catalan coast (Sarda and others 2005); thus, the they were selected because a large part of the socio-
accommodation coefficient was included as a relevant economic data available is only complete for 68.5% of
socioeconomic factor. The group of themes corre- municipalities (those with more than 5000 residents).
sponding to the environmental dimension coincided Themes were spatially combined using the GIS to
with the main institutional and governmental produce an ordinal pseudo-indicator of a specific
123
Environ Manage (2006) 38:993–1005 999
tion of integrated coastal zone management in
desirable condition of each theme. The resulting con-
Europe (EC 2002).
tinuous real number scale for each theme was numeri-
• They should constitute local administrative (man-
cally aggregated into an arbitrary four-way
agement) units.
classification, whether or not it was originally on an
• They should be based on real, natural, biophysical
ordinal scale. Gornitz (1990) and Gornitz and others
data.
(1994) used a similar approach to develop indices of
• They should integrate and reflect the principal
several coastal characteristics that were aggregated into
existing structure and functional processes of the
a vulnerability index using a linear model. The classifi-
coastal environment.
cation method used the Jenks optimization, which
• They should be derived from a combination of
identifies break points between classes by minimizing
independent characteristics that remain constant
the sum of the variance within each of the classes (Jenks
over time (wherever possible).
1967). This method identifies groupings and patterns
inherent in the data and produces a more objective
The natural and socioeconomic rationalizations
aggregate representation of spatial variability, thus
were aggregated to form the final HEMU map. The
providing a valuable tool with which to explore and
aggregation process obeyed certain algebraic combi-
represent data by minimizing its natural variation (Smith
nation rules. The final regional HEMU map was pro-
1986).
duced using four category units for the 12 comarcas of
the Catalan coast. An additional analytical phase de-
Data Aggregation Method fined spatial modeling rules to determine criteria for a
proposed natural coastal resources conservation
The natural themes considered in the analysis scenario.
(Table 1) were aggregated at the level of the comarca
to be spatially coherent and consistent with the socio-
economic data scale. An aggregation method based on
Results
a weighted average was used to represent the contri-
bution of the surface area of coastal municipalities to
The implementation of the Catalan ICZM strategic
the comarca level for the natural dimension themes
plan requires a territory-based spatial framework,
(see Gornitz 1990 for a discussion of data aggregation
which in this case is based on the definition of HEMUs.
methods). This met the requirement to establish a
Although the coastal system consists of several differ-
common spatial framework and prevented inferences
ent dimensions that determine its stability and health,
from higher to lower levels of analysis that are asso-
only two are used in this study: the socioeconomic and
ciated with the ecological fallacy (Alker 1969, Cao and
natural dimensions. It was assumed that the Catalan
Lam 1997). Comarcas constitute true physical man-
coastal zone could be defined for management pur-
agement units, because they are based on the common
poses in terms of these two dimensions, consisting of
historical, cultural, and administrative characteristics
five and six themes respectively that were incorporated
of their constituent municipalities. They are therefore
in the GIS at cartographic scales of 1:25,000 to 1:50,000
important in ICZM planning and monitoring of the
(Table 1).
´
Spanish coast (Barragan 2004).
Table 2 shows the values generated by classifying
Catalan coastal comarcas using the Jenks method for
Regionalization Process
each theme and dimension. This classification is based
on results given in terms of ordinal classes, where the
Theme typologies were used to develop a specific
maximum value (four) indicates the highest relevance
regionalization map for each dimension. The algebraic
of the characteristic and the minimum (one) indicates
sum of individual themes represented the contribution
the lowest relevance. Table 2 also shows the surface
of the individual natural and socioeconomic regional-
area (in hectares) of the comarcas and provinces to
ization of the Catalan coast. The thematic map of each
indicate the relative geographical contribution of the
dimension represents an independent view of the ter-
themes in the regionalization process.
ritory, and together they constitute the main input for
Figure 2 shows the results of the socioeconomic and
the integrated regionalization process. The following
the natural thematic rationalizations. There is a clear
criteria form the basis of the HEMU definitions:
relationship between the two: in general, higher values
• They should follow the principles proposed in the for the socioeconomic component are accompanied by
EU recommendation concerning the implementa- lower values for the natural component. This pattern
123
1000 Environ Manage (2006) 38:993–1005
Table 2 Theme classification values by comarca of the Catalan coastal zone
Socioeconomic Natural
Comarca Province Has A B C D E F G H I J K
`
Alt Emporda Girona 135,697 1 2 1 3 1 3 3 4 4 2 1
`
Baix Emporda Girona 70,016 1 2 1 3 2 2 2 3 3 2 1
Selva Girona 99,5374 1 3 1 4 1 2 1 3 3 2 1
Maresme Barcelona 40,049 2 3 2 2 2 1 2 3 3 2 3
`
Barcelones Barcelona 14,463 4 1 4 I 4 1 2 2 1 2 2
Baix Llobregat Barcelona 48,664 3 2 3 1 3 2 2 3 2 2 4
Garraf Barcelona 18,503 1 4 1 2 2 2 2 3 3 2 3
´
Baix Penedes Tarragona 29,618 1 4 1 2 2 1 1 3 2 1 3
`
Tarragones Tarragona 31,931 1 2 2 3 3 2 2 3 2 2 2
Baix Camp Tarragona 69,633 1 2 1 2 1 2 1 4 3 1 3
Baix Ebre Tarragona 100,212 1 2 1 1 1 3 2 4 4 3 4
`
Montsia Tarragona 73,741 1 2 1 1 3 3 3 4 3 4
Themes: (A) Population size; (B) Population growth; (C) Gross National Product; (D) Accommodation coefficient; (E) Impervious
surface; (F) Natural protected area; (G) Geomorphologic relevance; (H) Vegetation condition; (I) Landscape transformation;
(J) Running water condition; (K) Coastal geomorphology
Fig. 2 Socio-economic and natural
regionalizations of the Catalan coast
clearly reflects the central role of the metropolitan integrated measurement of the two dimensions. How-
`
areas of Barcelona (Barcelones) and Tarragona (Tar- ever, this method of aggregation can introduce inter-
`
ragones) in the socioeconomic development of Cata- pretation errors, because zones with very different
lonia. The least developed areas in socioeconomic characteristics can have similar numerical values. Thus,
` `
terms correspond to those with the highest environ- Barcelones and Alt Emporda have an equal total
mental values and are located in the northern (Alt wealth value which in the first case is due to a high
` `
Emporda) and southern (Montsia) ends of the region, socioeconomic value and in the second is due to a
where the most important protected natural coastal high natural one.
areas are located (Cap de Creus and the Ebre delta, To prevent this, we used an integrative model in
respectively; see Figure 1). which the natural component was combined with the
Once these two independent rationalizations were socioeconomic component, but in which they were
performed, they were combined to define the map of inverse scaled (i.e., an original value of four is substi-
the HEMUs. Figure 3 shows the HEMUs obtained by tuted by a value of one) and averaged. The resulting
applying a method designed to retain the attribute values were obtained from the algebraic mean of both
homogeneity of units after aggregation. By applying a the regionalization of the dimensions and re-aggre-
direct averaging of the two dimensions, the numerical gating them to their class type (i.e., values ranging from
values attached to each comarca in Figure 3 should be 2.000 to 2.999 indicate class 2). Reclassified values were
obtained. This value, which we will refer to as ‘‘total assigned to a nonordinal nominal four-class scheme to
wealth,’’ is obtained by averaging natural and socio- avoid misinterpretation of results. The final results
economic values, and it can be considered an indicate units (comarcas) with similar socioeconomic
123
Environ Manage (2006) 38:993–1005 1001
Fig. 3 Homogeneous Environmental Management Units of the Fig. 4 Conservation HEMU regionalization scenario of the
Catalan coast. Numbers indicate total socio-economic and Catalan coast
natural total richness by unit
for management purposes. As in previous cases, the
and natural properties but without showing any prior- maximum value for the criteria selected is four; in this
case, the highest environmental values. Management
ity indication.
The four-class comarca map obtained represents a plans for these units should be properly considered.
reliable management regionalization of the Catalan
coast, while being a data-based and user-oriented
product. Based on the spatial aggregation method Discussion and Conclusions
developed, it was possible to account for the functional
homogeneity of the coastal zone. The HEMUs classify The GIS provided an appropriate geospatial structure
the comarcas into highly natural areas (A), seminatural through which to develop an efficient classification of
areas (B), semiurban areas (C), and areas with high coastal management units (Shupeng 1988). As sug-
socioeconomic development (D). Geographically, each gested by Bartlett (2000), GIS also played a key role in
of these classes (units) should be managed under a database construction, theme modeling, and visualiza-
desired ‘‘vision’’ that fulfills the expectations of the tion of results. Although the selected polygon data
population living in the area and obeys the established model does not account for the dimensional problems
legal framework. implicit in the line representation of the coast (Vafeidis
Finally, the need to incorporate a stronger plan for and others 2004), we also found that there is no
the conservation of natural resources in current and straightforward system to define an aerial model that
future coastal zone management strategies has been efficiently manages the dynamics of the two coastal
stressed previously by several authors (Sherman and dimensions studied (Mueller and others 2002). How-
Alexander 1986, Van der Weide 1993, EC 2002, ever, in this study we used the mean-based aggregation
DMAH 2004). A management scenario involving model proposed by Gornitz (1990), because it has been
environmental conservation was defined to conserve demonstrated to be less sensitive to data errors, omis-
the natural role of the coast and provide a tool for sions, and misclassifications.
managers that could contribute to the target set for In order to use a method that is general enough to
2010 by the Convention on Biological Diversity (CBD be applicable to most coastal zones, themes describing
1999). The scenario was defined by applying an arbi- each component were selected on the basis that they
trary relative weighting of 80% to the natural dimen- were relevant, georeferenced, and could be either
sion values and 20% to the socioeconomic values. easily measured or obtained from existing official data.
Figure 4 shows the resulting map of HEMUs in terms Although it might be desirable to integrate data at a
of ordinal values. In this case, the map represents con- larger cartographic scale, positive results were ob-
servation priorities for the Catalan coast. The regions served in the spatial patterns obtained at the Catalan
are clearly similar to those obtained from the equally coast geographic scale. This is clearly the result of the
weighted averaging map, with the differences between multisource database appropriate integration at a
them arising from the existence of priority indications subnational cartographic scale (1:25,000–1:50,000), as
123
1002 Environ Manage (2006) 38:993–1005
The integrated description of themes selected for
recommended by UNEP (1995) (Table 1). The themes
the Catalan coastal zone can be considered represen-
are relevant to most developed and developing coastal
tative of developed coastal areas, where high values for
zones, and only a few were specific to the coast ana-
the socioeconomic components are frequently accom-
lyzed. This approach differs from data-intensive studies
panied by low values for the natural components
requiring a large number of descriptors for each theme
(Figure 2) This also seems to confirm a global tendency
that in many cases prevent its practical application (see
in coastal areas for socioeconomic activities to gener-
an example in Cendrero and Fischer 1997). An exam-
ate significant pressures on coastal systems, leading to
ple of an area-specific variable is the accommodation
an inherent reduction in or degradation of natural re-
coefficient (number of hotel beds per inhabitant),
sources. A similar pattern was found in comarcas with
which is only relevant to areas in which tourism is an
high values for natural resources (the northern and
important economic activity. This is clearly the case for
southernmost comarcas); although these are the least
the Catalan coast, where tourism accounted for about
developed in socioeconomic terms, they were the
10.8% of GNP in 2002 (DCTC 2002). If this analysis
greatest contributors to the geographic and biological
were to be performed for a coastal zone with different
diversity of the Catalan coast. If a river subbasin
major economic sectors, the corresponding represen-
schema existed for this area, the present results could
tative indicator would need to be properly selected to
be complemented in the future with similar approaches
reflect the most important socioeconomic component.
´
to those used by Escofet (2002) and Yanez-Arancibia
In this study, two parameters in the natural dimension
and Day (2004).
were calculated for the 200-m-wide fringe along the
Whenever the natural and socioeconomic dimen-
coast using the GIS, instead of using municipalities as
sions have to be integrated in order to characterize the
spatial units. This was done to accurately reflect coastal
properties of discrete planning units, results can be
environmental resources and not environmental re-
unclear or susceptible to misinterpretation by manag-
sources in coastal administrative units (municipalities)
ers. This is due to the inverse relationship between the
in a specifically adapted ecosystem approach (Rappa-
socioeconomic and natural values of developed coastal
port 1999). This width corresponds to the official con-
areas mentioned above. Thus, two units with different
servation easement zone based on the administrative
characteristics (one with high socioeconomic and low
regulations for the Spanish coast (Spanish Coastal Law,
natural values, and the other with the reverse situation)
BOE 1989) and must therefore be adjusted to the spe-
could give the same overall integrated value if they
cific regulations of the coast to be analyzed.
were directly combined. Although the value obtained
The natural data layers at the level of the munici-
in this way could be interpreted as a measure of the
pality were aggregated at a higher administrative
total wealth (considering both themes) of the territo-
level—the comarca—by considering values corre-
rial unit, it is clear that the two units could not be
sponding to the number of coastal municipalities in-
managed in the same way. This problem was overcome
cluded in it. Thus, the use of comarcas, made up of
by reclassifying one of the components before adding
municipalities with similar characteristics, leads to a
them together and prevented the socioeconomic data
degree of uniformity that is most likely to be due to the
interval ranking problems experienced by McLaughlin
common natural and socioeconomic environment that
and others (2002). The implicit result of this operation
is implicit inside the boundaries, reflected in a unifica-
should be equivalent to only considering one of the two
tion effect within the comarca. This final geographic
components, and it can only be used for coasts that
scale was found to be useful for reducing the high var-
display the inverse relationship between socioeco-
iability found at the level of municipalities, which would
nomic and natural values mentioned above.
have complicated the design of an effective ICZM
The bottom-up approach used here provided a data-
strategic plan for Catalonia (or probably anywhere
driven environmental regionalization of the coast that
else). This scale still retained the major sources of var-
could not have been obtained with a predetermined
iability along the coast, and because data were upscaled
planning structure (Harff and Davis 1990). Thus, the
and no aerial subdivisions were made, it did not show
results obtained are not intended to provide a priori
significant MAUP symptoms (Marceau 1999). Like-
management priorities, but rather to identify classes of
wise, no scale-dependent problems were addressed in
truly homogeneous units that managers can use for
the classification process because several themes were
future planning, policy implementations, and moni-
compiled from the beginning at the comarca level and
toring initiatives. This can be seen clearly in Figure 3
not aggregated at a different resolution. Similar results
by comparing the difference in HEMU class (four
concerning the use of comarcas as aggregation planning
´ classes) with the total wealth values obtained (almost
units in Spain can be found in Barragan (2004).
123
Environ Manage (2006) 38:993–1005 1003
constant throughout the entire territory). However,
HEMUs with the lowest total wealth values (La Selva,
`
Maresme, Baix Penedes, and Baix Camp) should be
identified as critical hot spots in the ICZM strategic
plan. Compared with the rest of the territory, these hot
spots do not seem to have a dominant value or re-
source. As suggested by Burbridge (1997), a special
plan would have to be designed to improve their situ-
ation and to reach the average value throughout the
territory.
Following the recommendations of the Sixth Envi-
ronmental Action Programme of the European Com-
munity (EC), the conservation of natural resources has
been defined as a central objective of the Catalonia
ICZM strategic plan to maintain and/or improve the
environmental quality of the system and its associated Fig. 5 Touristic regionalization of the Catalan coast (DCTC
2002)
human societies (DMAH 2004). The specific conser-
vation regionalization developed here (Figure 4) pro-
vides a spatial vision based on the natural quality of the fails to reflect the actual socioeconomic and natural
coastal zone and at the same time serves to identify variability and complexity of the coastal zone. This
priority conservation areas, a process that has been generates five regions (Figure 5) that, despite being
proposed as relevant to coastal management by EC currently managed and exploited as homogeneous
(2002). According to the pattern observed, the areas units, are composed of comarcas with dissimilar
with the highest environmental values are the north- socioeconomic and natural values (Figures 2 and 3).
ernmost and southernmost comarcas, and conse- The method proposed here to define a multidimen-
quently, under the present management scenario, those sional HEMU-based regionalization of the coastal
are the areas with the lowest priority. For the comarcas zone using GIS overcomes these problems and can be
with the lowest natural values, two different manage- used to define a more integrated management plan.
ment options could be selected: defining immediate However, the present proposal represents the result of
actions for the improvement of environmental values a data-driven analysis, and the process should be
(condition) or abandoning them and converting them complemented by a more social vision that considers
into sacrificed areas in terms of natural wealth. The final the goals and interests of managers, stakeholders, and
choice will depend on the level of transformation shown end users.
by these areas, as well as local institutional capacity. In In summary, the regionalization process performed
any case, to build a management-oriented scenario, the here for the Catalan coastal zone generated four dif-
selection of weights for the socioeconomic and natural ferent classes of HEMUs, for which socioeconomic and
components should be based on real policy objectives natural characteristics were combined in a GIS to give
as part of a more systemic view (Van der Weide 1993). an overall integrated value. The GIS proved to be an
Thus, this study only represents a proposal for manag- efficient tool for data management, analysis, and visu-
ers to consider in relation to such issues. alization in the overall process of defining coastal
Although based on the comarca administrative management units. This HEMU-based regionalization
units, the regionalization of the Catalonia coastal zone of the territory is a way to rationalize the definition of
based on HEMUs performed here does not correspond the Catalan ICZM strategic plan. This geospatial ap-
to any other existing comarca-based regionalization of proach could also be adapted and applied to other
the area. Most existing regionalizations are based on a coastal regions. Finally, the relevance of the process
single theme (typology) and consequently fail to cap- will ultimately depend on specific management goals
ture the integrated structure and functioning of the and objectives, and must be considered in the context
coastal system. As an example, the Catalan coastal of the need for a multicomponent spatial vision of the
tourism regionalization (DCTC 2002) is based on the coastal system. The proposed HEMU regionalization,
major economic driving force for the coastal zone, e.g., based on the comarca as the administrative/manage-
the tourist industry. In spite of the relative weight of ment unit, is expected to be an important tool for the
this factor in the socioeconomic structure, using it as future implementation of the recent ICZM strategic
the only regionalization parameter for the territory plan for Catalonia.
123
1004 Environ Manage (2006) 38:993–1005
Acknowledgments This work has been carried out within the Cendrero A., D. W. Fischer. 1997. A procedure for assessing the
framework of the Mevaplaya project (REN2003-09029-C03-01/ environmental quality of coastal areas for planning and
MAR), which is funded by the Spanish Ministry of Education management. J Coastal Res 13:732–744
and Science. The first author was supported by a doctoral grant Christian C. S. 1958. The concept of land units and land systems.
´
from the National Science and Technology Council of Mexico Proc Ninth Pacific Congress 20:74–81
(CONACyT) and the second author by the University Research DCTC. 2002. Catalonia tourism in numbers 2002. Tourism studies of
Promotion Award for Young Researchers of the Government of Catalonia. Commerce, Tourism and Consumer Department
Catalonia. The authors wish to thank the organizations and (DCTC), Autonomous Government of Catalonia, Barcelona,
institutions that supplied the data used in this work. Special Spain, 8 pp
´
thanks are due to Albert Ferre (Universitat de Barcelona), DMAH. 2003. Marine accidental pollution emergency special
´
Ramon Jordana (Departament de Agricultura, Ramaderia plan of Catalonia (CAMCAT). Environment and Housing
`
i Pesca, Generalitat de Catalunya), Marta Manzanera (Agencia Department (DMAH), Autonomous Government of Cata-
´
Catalana de l’Aigua), and Xavier Martı (Departament de Medi lonia, Barcelona, Spain (CD-ROM)
Ambient i Habitatge, Generalitat de Catalunya). We are also DMAH. 2004. Integrated coastal zone management strategic
`
grateful to Modest Fluvia (Universitat de Girona) for his help in plan of Catalonia (PEGICZ). Environmental Section.
defining the economic dimension. We thank Y. Henocque and Environment and Housing Department (DMAH), Auton-
two other anonymous reviewers for their comments and sug- omous Government of Catalonia, Barcelona, Spain, 66 pp
gestions on the original manuscript. DPTOP. 1983. Ecological and environmental study of coastal
sections. In ports and Transportation General Direction,
Territorial Policy and Public Works Department (DPTOP)
(ed.) Recreational Ports Plan Study. Volume 2. Autonomous
References
Government of Catalonia, Barcelona, Spain, pp 273–390
EC. 2000. Council Directive of the European Parliament and the
Alker R. 1969. A typology of ecological fallacies. In Dogan M, Council of 23 October 2000. Establishing a framework for
M. Rokkan (eds), Quantitative ecological analysis in the community action in the field of water policy. Official
social sciences. MIT Press, Cambridge, Massachusetts, pp 3 Journal of the European Communities (2000/60/EC), Eur-
Amir S. 1987. Classification of coastal resources: a Mediterra- opean Commission (EC), Brussels, Belgium, 72 pp
nean case study. Landscape Urban Planning 14:399–414 EC. 2002. Recommendation of the European Parliament and of
Baja S., D. M. Chapman, D. Dragovich. 2002. A conceptual the council of 30 May 2002 concerning the implementation
model for defining and assessing land management units of integrated coastal zone management in Europe. Official
using a fuzzy modelling approach in GIS environment. En- Journal of the European Communities (2002/413/EC),
vir Manage 29:647–661 European Commission (EC), Brussels, Belgium, 4 pp
´
Barragan J. M. 2004. The littoral areas of Spain. From geo- EEA. 2003. DMEER: Digital Map of European Ecological Re-
graphic analysis to integrated management. Ariel S.A., gions. The European Topic Centre on Nature Protection
Barcelona, Spain, 214 pp. and Biodiversity. European Environmental Agency [online:
Bartlett D. 2000. Working on the frontiers of science: applying http://dataservice.eea.eu.int/atlas/viewdata/viewpub.asp?id=7],
GIS to the coastal zone. In Wright D, D. Bartlett (eds), revised on 20 April 2005
Marine and coastal geographical information systems. Tay- ´
Escofet A. 2002. Alternativas para la regionalizacion del espacio
lor & Francis, London, United Kingdom, pp 11–24 ´
marino de Mexico. Working document prepared for the
Bartley J. A., J. W. Buddemeier, D. A. Bennett. 2001. Coastline Mapping Marine and Estuarine Ecosystems of North
complexity: a parameter for functional classification of ´ ´
America Project. Centro de Investigacion Cientıfica y de
coastal environments. J Sea Res 46:87–97 ´
Educacion Superior de Ensenada & Commission for Envi-
Belfiore S. 2003. The growth of integrated coastal management ronmental Cooperation, NAFTA, Ensenada, Mexico, 13 pp
and the role of indicators in integrated coastal management: Finkl C. W. 2004. Coastal classification: systematic approaches to
introduction to the special issue. Ocean Coastal Manage consider in the development of a comprehensive scheme.
46:225–234 J Coast Res 20:166–213
Bian L. 1997. Multiscale nature of spatial data in scaling up Fricker A., D. L. Forbes. 1988. A system for coastal description
environmental models. In Quattrochi D. A, M. F. Goodchild and classification. Coast Manage 16:111–137
(eds), Scale in remote sensing and GIS. Lewis Publishers, Goodchild M. F. 2000. Foreword. In Wright D, D. Bartlett (eds),
Boca Raton, Florida, pp 13–26 Marine and coastal Geographical Information Systems.
BOE. 1979. Organic Law 4/1979, of December 18, Catalonia Taylor & Francis, London, United Kingdom, pp xiii–xv
Autonomous Statute. State Official Newsletter (BOE), Gornitz V. 1990. Vulnerability of the East Coast, U.S.A. to fu-
Published 22 December 1979, Madrid, Spain ture sea level rise. J Coast Res special issue no. 9:201–237
BOE. 1989. Law 22/1988, of July 29 (Leadership of the State) of Gornitz V. M., R. C. Daniels, T. W. Whites, K. R. Birdwell. 1994.
Coasts. State Official Newsletter (BOE), Legal Documents The development of a coastal risk assessment database:
Collection. Published January 1994, Madrid, Spain, 389 pp vulnerability to sea-level rise in the U.S. Southeast. J Coast
Burbridge P. R. 1997. A generic framework for measuring suc- Res special issue no. 12 (Coastal Hazards):327–338
cess in integrated coastal management. Ocean Coastal Harff J., J. C. Davis. 1990. Regionalization in geology by mul-
Manage 37:175–189 tivariate classification. Math Geol 22:573–588
Cao C., Lam N. 1997. Understanding the scale and resolution Henocque Y. 2003. Development of progress indicators for
effects in remote sensing and GIS. In Quattrochi D. A., M. coastal zone management in France. Ocean Coast Manage
F. Goodchild (eds), Scale in remote sensing and GIS. Lewis 46:363–379
Publishers, Boca Raton, Florida, pp 57–72 Henocque Y., B. Andral. 2003. The French approach to man-
CBD. 1999. Liaison group on the ecosystem approach. Con- aging water resources in the Mediterranean and the new
vention on Biological Diversity & UNESCO, Workshop European Water Framework Directive. Marine Pollut Bull
Report, September 15–17, 1999, Paris, 11 pp 47:155–161
123
Environ Manage (2006) 38:993–1005 1005
IDESCAT. 2005. 2005 Book of statistics. Statistics Institute of SDAGE, 295 pp. Volume 3: Cartography objectives and
Catalonia [online: http://www.idescat.net/], revised 22 April priorities, 15 A3 maps. Rhone-Mediterranean Watershed
2005 Delegation (RMC-Comite de Bassin), France
´
Jenks G. F. 1967. The data model concept in statistical mapping. Sarda R., C. Avila, J. Mora. 2005. A methodological approach to
Int Yearbook Cartogr 7:186–190 be used in integrated coastal zone management process: the
Marceau D. J. 1999. The scale issue in social and natural sciences. case of the Catalan Coast (Catalonia, Spain). Estuarine
Can J Remote Sens 25:347–356 Coastal Shelf Sci 62:427–439
Maxwell B. A., R. W. Buddemeier. 2002. Coastal typology Sherman K., L. M. Alexander. 1986. Variability and manage-
development with heterogeneous data sets. Regional Envir ment of large marine ecosystems. AAAS selected sympo-
Change 3:77–87 sium 99. Westview Press, Boulder, Colorado, 319 pp
McLaughlin S., J. McKenna, J. A. G. Cooper. 2002. Socio-eco- Shupeng C. 1988. The coastline as a base for global databases: a
nomic data in coastal vulnerability indices: constrains and pilot study in China. In Tomlinson R (ed), Building data-
opportunities. J Coast Res 36:487–497 bases for global science. Taylor & Francis, London, United
Mee L. D. 2005. Asssessment and monitoring requirements for Kingdom, pp 202–215
the adaptive management of Europe’s Regional Seas. In Smith R. M. 1986. Comparing traditional methods for selecting
Salomons W, Vermaat J, K. Turner (eds), Managing Euro- class intervals on choropleth maps. Professional Geographer
pean coasts: past, present and future. Environmental Sci- 38:62–67
ences Series, Springer-Verlag, Berlin, Germany, pp 227–237 UNEP. 1995. The development and implementation of ICAM.
Meentemeyer V., E. O. Box. 1987. Scale effects in landscape In UNEP regional seas reports and studies (ed), Guidelines
studies. In M. G. Turner (ed), Landscape heterogeneity and for integrated management of coastal and marine areas:
disturbance. Springer-Verlag, New York, New York, pp 16–34 with special reference to the Mediterranean basin. PAP/
Mueller M., B. Meissner, W. Weinrebe. 2002. TerraMarIS–Ter- RAC (MAP-UNEP), no. 161, Split, Croatia, pp 19–33
restrial and Marine Information System. In Breman J (ed), UNESCO. 1997. Definition of the coherent management units:
Marine geography: GIS for the oceans and the seas. ESRI Stage 2. In Methodological guide to integrated coastal zone
Press, Redlands, California, pp 92–102 management. Manuals & guides 36. Intergovernmental
´ ´
Nogue J. 2004. The territorial and landscape transformation of Oceanographic Comision, France, pp 16–19
the Costa Brava (1956-2003): Present situation and pro- Vafeidis A. T., R. J. Nicholls, L. McFadden, J. Hinkel, P. S.
posed activities. In workshop 1: Scarce Territory or Fragile Grashoff. 2004. Developing a global database for coastal
landscape. Costa Brava Debate (March 5, 2004) [online: vulnerability analysis: design issues and challenges. In XX
http://www.debatcostabrava.org/], revised on 15 November ISPRS Congress (ed), The international archives of the
2005. Rosas, Spain photogrammetry, remote sensing and spatial information
Nunneri C., K. R. Turner, A. Cieslak, A. Kannen, R. Klein, L. sciences, vol. 34, part XXXV, commission IV, 12–23 July,
Ledoux, J. Marquenie, L. Mee, S. Moncheva, R. Nicholls, Istanbul, Turkey, pp 801–805
´
W. Salomons, R. Sarda, M. Stive, T. Vellinga. 2005. Inte- Van der Weide J. 1993. A systems view of integrated coastal
grated assessment and future scenarios for the coast. In management. Ocean Coast Manage 21:129–148
Vermaat J, L. Bouwer, K. Turner, W. Salomons (eds), Walpole S. C. 1998. Integration of economic and biophysical
Managing European coasts: past, present and future. Envi- information to assess the site-specific profitability of land
ronmental Sciences Series, Springer-Verlag, Berlin, Ger- management programmes using a GIS. In Proceedings of
many, pp 271–290 the Eighth ISCO Conference, New Delhi, India, pp 1663–
Openshaw S. 1984. The modifiable aerial unit problem. CAT- 1669
´˜
MOG 38. Geobooks, Norwich, United Kingdom Yanez-Arancibia A., J. W. Day. 2004. Environmental sub-
Rappaport J. 1999. The ecosystem approach from a practical regions in the Gulf of Mexico coastal zone: the ecosystem
point of view. Conserv Biol 13:679–681 approach as an integrated management tool. Ocean Coast
RMC-Comite de Bassin. 1995. Zoning and management scheme Manage 47:727–757
of the Rhone-Mediterranean and Corsica basin (SDAGE). Zonneveld I. S. 1994. Basic principles of classification. In Klijn F
Users’ guide, 13 pp. Volume 1: Key fundamental orienta- (ed), Ecosystem classification for environmental manage-
tions, operational measures and modalities, 120, pp: Volume ment. Kluwer Academic Publishers, The Netherlands,
2: Thematic archives, rules and recommendations of pp 23–47
123