See

discussions, stats, and author profiles for this publication at: https://www.researchgate.net/publication/299475959

An index-based method to assess risks of

climate-related hazards in coastal zones: The
case of Tetouan

Article in Estuarine Coastal and Shelf Science · March 2016

DOI: 10.1016/j.ecss.2016.03.021

CITATIONS READS

6 425

5 authors, including:

Alessio Satta Maria Snoussi

Mediterranean Sea and Coast Foundation (M… University Mohammed V
20 PUBLICATIONS 23 CITATIONS 50 PUBLICATIONS 510 CITATIONS

SEE PROFILE SEE PROFILE

Manuela Puddu Hout Radouane

Mediterranean Sea and Coast Foundation (M… University of Limoges
5 PUBLICATIONS 8 CITATIONS 1 PUBLICATION 6 CITATIONS

SEE PROFILE SEE PROFILE

Some of the authors of this publication are also working on these related projects:

CAMP Italy View project

PHC VOLUBILIS MA/11/253 View project

All content following this page was uploaded by Maria Snoussi on 10 October 2017.

The user has requested enhancement of the downloaded file.

 Estuarine, Coastal and Shelf Science 175 (2016) 93e105

Contents lists available at ScienceDirect

Estuarine, Coastal and Shelf Science

journal homepage: www.elsevier.com/locate/ecss

An index-based method to assess risks of climate-related hazards in

coastal zones: The case of Tetouan
Alessio Satta a, *, Maria Snoussi b, Manuela Puddu a, Latifa Flayou b, Radouane Hout b
a
University of Cagliari, DICAAR e Department of Civil and Environmental Engineering and Architecture, via Marengo 2, 09123 Cagliari, Italy
b
University Mohamed V, Faculty of Sciences, Department of Earth Sciences, Avenue Ibn Battota, BP 1014, Rabat, Morocco

a r t i c l e i n f o a b s t r a c t

Article history: The regional risk assessment carried out within the ClimVar & ICZM Project identified the coastal zone of
Received 25 September 2015 Tetouan as a hotspot of the Mediterranean Moroccan coast and so it was chosen for the application of the
Received in revised form Multi-Scale Coastal Risk Index for Local Scale (CRI-LS). The local scale approach provides a useful tool for
16 February 2016
local coastal planning and management by exploring the effects and the extensions of the hazards and
Accepted 26 March 2016
Available online 28 March 2016
combining hazard, vulnerability and exposure variables in order to identify areas where the risk is
relatively high. The coast of Tetouan is one of the coastal areas that have been most rapidly and densely
urbanized in Morocco and it is characterized by an erosive shoreline. Local authorities are facing the
Keywords:
Sea-level rise
complex task of balancing development and managing coastal risks, especially coastal erosion and
Morocco flooding, and then be prepared to the unavoidable impacts of climate change. The first phase of the
Coastal zone application of the CRI-LS methodology to Tetouan consisted of defining the coastal hazard zone, which
Risk assessment results from the overlaying of the erosion hazard zone and the flooding hazard zone. Nineteen variables
Index-based method were chosen to describe the Hazards, Vulnerability and Exposure factors. The scores corresponding to
each variable were calculated and the weights assigned through an expert judgement elicitation. The
resulting values are hosted in a geographic information system (GIS) platform that enables the individual
variables and aggregated risk scores to be color-coded and mapped across the coastal hazard zone. The
results indicated that 10% and 27% of investigated littoral fall under respectively very high and high
vulnerability because of combination of high erosion rates with high capital land use. The risk map
showed that some areas, especially the flood plains of Restinga, Smir and Martil-Alila, with distances
over 5 km from the coast, are characterized by high levels of risk due to the low topography of the flood
plains and to the high values of exposure. The CRI-LS provides a set of maps that allow identifying areas
within the coastal hazard zone with relative higher risk from climate-related hazards. The method can be
used to support coastal planning and management process in selecting the most suitable adaptation
measures.
© 2016 Elsevier Ltd. All rights reserved.

1. Introduction
Mediterranean coastal zones support a very high population
density that leads to high social and bio-geophysical vulnerabilities
as coastal infrastructures are exposed to direct waves and the lack
of space for these vulnerable systems to be moved to less vulner-
able areas. The increased risks of natural hazards generated and/or
exacerbated by sea level rise and marine storms and the growing
concentration of people and activities on the coastal zone, requires
updated information and a better understanding on coastal zones
* Corresponding author. Present address: Via Giotto 6, 09121 Cagliari, Italy.
E-mail address: satta.alessio@gmail.com (A. Satta).
vulnerabilities and risks at the local scale (Satta, 2014). Even if
extreme events often cannot be predicted, adaptation measures can
be planned to reduce the potential risks and to cope with un-
certainties. Notwithstanding these emerging risks, lack of robust
scientific knowledge, lack of local data, local experts have led to
coastal decision makers under-evaluating sea level rise as an im-
€
mediate threat (Ozyurt and Ergin, 2010). These uncertainties de-
mand a high degree of flexibility to adapt to climate and non-
climate driven changes and, in this sense, designing and imple-
menting a method to assess current and future risks to coastal
hazards is a challenging issue for research.
These issues are even more acute in some coastal areas of the
southern shore of the Mediterranean, where local reliable data are
often lacking and where local capacities to conduct risk assessment

http://dx.doi.org/10.1016/j.ecss.2016.03.021
0272-7714/© 2016 Elsevier Ltd. All rights reserved.
94 A. Satta et al. / Estuarine, Coastal and Shelf Science 175 (2016) 93e105
need to be strengthened. Therefore the development of a simple
and cost-effective method to assess vulnerability and risk, which
allows coastal managers and planners to rapidly determine high-
risk areas in the coastal zone, is crucial to manage the coastal
zone effectively over the next decades.
In scientific literature several models and tools to assess coastal
vulnerability and risk exist that differ in complexity, in the number
of processes that they include, the application at various scales and
in their outputs. Just few comprehensive reviews were made to
assist coastal managers in the selection of an appropriate method
for conducting a coastal vulnerability assessment (Mcleod et al.,
2010; ETC-CCA, 2011; Satta, 2014).
Satta (2014) prepared a review and analysis of existing methods
for the assessment of coastal vulnerability and risk to climate and
non-climate induced hazards at different spatial scale (from
regional to local). In this work, 26 tools were selected and then
clustered in four main categories (Index-based tools, tools based on
dynamic computer models, GIS Based Decision Support tools,
Visualization tools). The tools must meet three main criteria
(including climate and non climate drivers and related impacts;
suitable with the conditions of a local coastal community; outputs
useful to support local adaptation), which in turn are divided into
nine requirements (Fig. 1). These requirements intend to describe
the needs of a Mediterranean coastal community in terms of
assessment to current and future risks to coastal hazards (Satta,
2014).
Most of the vulnerability and risk assessment studies, developed
so far, are primarily driven by top-down processes in which local
managers and other local stakeholders are not involved (Satta,
2014). Moreover most existing vulnerability and risk assessment
methods were not developed starting from the needs of the local
communities (Satta, 2014). Another important limitation is that
adaptive capacity and response of systems to climate change
drivers and related impacts are often unknown (Mcleod et al.,
2010). These aspects represent a concrete weakness common to
several existing tools. Nevertheless, four methods were identified
€
as meeting these nine requirements: CVI-SLR (Ozyurt, 2007), Multi-
scale CVI (McLaughlin and Cooper, 2010), DESYCO (Torresan et al.,
2010), SimCLIM (CLIMsystems Ltd). The Index-based methods
(CVI - SLR and Multiscale e CVI) present several advantages like to
be easily upgradeable (variables can be added or eliminated) and
algorithms employed for the calculation are readily understandable
for coastal managers and practitioners. DESYCO, developed by
Torresan et al. (2010) consists of identifying vulnerability indicators
and indices for the evaluation of climate change impacts in coastal
zones and it aims to identify and prioritize areas and targets at risk
in the considered region (ETC-CCA, 2011). DESYCO implements a
Regional Risk Assessment (RRA) methodology and it is based on
Multi-Criteria Decision Analysis (MCDA). The main limitation of
DESYCO, as reported by ETC-CCA (2011), consists in the potential
geographical errors induced by the diversity of data sources for-
mats and spatial scales. Among the methods based on dynamic
computer models, SimCLIM is very flexible, and can be customized
to local conditions (climatic, physical and socio-economical).
Indeed, its strengths are the ‘open-framework’ features that allow
its use in very different geographical and spatial conditions, which
meet the requirements, set by Satta (2014). However, as all the
dynamic computer models, the main limitation consists in the
upgradability of the algorithms and variables, which requires me-
dium to high expertise for its customisation to new regions (ETC-
CCA, 2011). A fifth method, which meets the nine requirements,
is the Regional Vulnerability Assessment (RVA) methodology
developed by Torresan et al. (2012). The RVA is an index-based
methodology with a heterogeneous subset of bio-geophysical and
socio-economic vulnerability indicators (Torresan et al., 2012). The
only limitation of this method to the objectives of this work, is that
its theoretical framework refers to the concept of vulnerability
defined in AR4 (IPCC, 2007).
Based on the analysis of main advantages and disadvantages
(Mcleod et al., 2010; ETC-CCA, 2011; Satta, 2014) the following
conclusions can be drawn in relation to the possible use of these
methods to assess coastal risk to climate change at the Mediter-
ranean local level:

methods refer to vulnerability concept as mainly incardinated in
the theoretical framework of IPCC (2007). All these studies have
been carried out before the definition of a new conceptual
framework for coastal risk as proposed by IPCC (2014a; 2014b);

most of these methods does not take into account the feedback
loops that operate between different subsystems and process at
various scales, which is a limitation of classical approaches
based on the sectorial analysis of the ecological or socio-
economical components (Satta, 2014);

just few methods address coastal risk at an appropriate scale
showing low flexibility to operating at various spatial scales;

several methods attempt to provide absolute predictions about
the impacts of climate change rather, than providing informa-
tion about the coastal areas likely to be affected more severely
than others (Torresan et al., 2010).
To overcome these limitations a method, called Multi-scale
coastal risk index (MS-CRI), was developed by Satta et al. (2015).
The conceptual framework adopted in the MS-CRI was initially
defined by Satta (2014). The MS-CRI was implemented and tested
within the framework of the ClimVar Project1 to carry out the
regional risk assessment of eleven Mediterranean countries and the
local risk assessment in one Mediterranean coastal spot. The overall
goal of the Climvar project, funded by GEF and coordinated by
UNEP, was to promote the use of ICZM in the participating Medi-
terranean countries as an effective tool to deal with the impacts of
climate variability and change in coastal zones by mainstreaming
them into the ICZM process.
The MS-CRI was deemed to be the method that best fitted the
purpose of the ClimVar project, due to the following characteristics:
it is a multi-scale risk assessment approach; integrates the theo-
retical framework of AR5 (IPCC, 2014a); integrates a large set of
socio-economic variables; risk assessment targets are represented
by variables that are separated into three sub-indices; takes into
account the interaction among different subsystems, presents an
easy calculation process to analyze the different vulnerability fac-
tors; and outcomes consist of vulnerability and risk maps. The MS-
CRI combines multiple variable layers, representing different as-
pects of risk, in sub-indexes (hazard, vulnerability and exposure)
layers, in such a way that risk ‘hotspots’, as well as areas of rela-
tively lower risk, emerge from the integration of the layers. The
application of the MS-CRI at the local scale (CRI-LS) aims to support
policy makers and coastal managers to evaluate how climate and
non-climate forcing interact with existing hazards to impact the
coastal zones. As direct policy implication the CRI-LS provides the
definition of the coastal hazard zone, under the 2100 scenario, and
a set of risk maps that can assist policy makers to prioritize coastal
management efforts that need to be undertaken to minimize risks
or mitigate the consequences of climate and non-climate related
hazards. Such a tool can be easily integrated into an overall coastal
management and adaptation strategies to support the imple-
mentation of the ICZM Protocol. To test Multi-Scale Coastal Risk
1
http://planbleu.org/en/activites/changement-climatique/climvar-project.
 A. Satta et al. / Estuarine, Coastal and Shelf Science 175 (2016) 93e105 95

Fig. 1. Requirements for the coastal vulnerability and risk assessment methods. Source: Satta (2014).

Index application at the Local Scale (CRI-LS) a case study on the
coastal zone of Tetouan in the Mediterranean coast of Morocco was
performed.
2. Description of the study area
Tetouan coast stretches over 40 km along the northwestern
Mediterranean coast of Morocco, between Fnideq village in the
north and Ras Mazari headland in the south (Fig. 2). The primary
shore types of the area are sandy beaches bordering small coastal
plains, with few coastal dunes and marshes. This long stretch is
interrupted only by the rocky cape of Cabo Negro, which rises to
332 m and drops almost vertically near the coast. Dune ridges, fossil
cliffs and a rocky shore platform are also present locally along the
studied area. The hydrodynamic conditions are characterized by
dominant waves from the E to ENE, a low tidal range and a principal
longshore drift northward (El Mrini, 2011). The coastal area is very
little fed with fluvial inputs (LPEE, 1997; Ibrahimi, 2003). Six more
or less small rivers (wadis) flow into the coast; namely from north
to south: Negro, Smir, Fnideq, Martil, Mellah and Azla, with Wadi
Martil, whose catchment has an area of 1220 Km2 being the most
important sedimentary source. The hydrological regime of these
watercourses is typically Mediterranean with flash floods during
the short-time high waters and an almost drying off in low waters.
The sediment fluxes are not significant especially after the con-
struction of dams on the Smir and Martil rivers, the most important
sediment sources. Besides that, no significant sedimentary inputs
are derived from cliff retreat processes (Niazi, 2007). In this context
96 A. Satta et al. / Estuarine, Coastal and Shelf Science 175 (2016) 93e105
Fig. 2. The coastal area of Tetouan.
of sediment deficit and since the last two decades, the coast has
been subjected to increasing human pressures, including residen-
tial construction on fore dunes and on vulnerable low-lying coasts,
construction of leisure ports (M'diq, Smir Marina and Kabila) and a
road network along the coast. Over 95% of the coastal dunes have
been destroyed by housing and tourism infrastructure. These
coastal developments have completely disturbed the cross shore
and/or interrupted the longshore sediment transport, leading to a
severe coastal erosion. In fact, Tetouan coast is one of the Medi-
terranean coastal areas that have been the most rapidly and densely
urbanized in Morocco. The area includes three cities (Martil, M'diq
and Fnideq) and several small villages. In 2010, almost 180,000
people resided permanently in the area with an average growth
rate projected to reach up to 7% over the period 2012e2015
(Benavente et al., 2007; Nachite, 2009) whilst the influx of tourists
can double during the summer months. Unfortunately, develop-
ment has been expanded without any integrated vision or long-
term planning taking into account the carrying capacity of the
coast. Highly developed sections of the coast coupled with the
high-energy, swell-dominated nature of the near shore makes it
increasingly vulnerable to coastal erosion, storm surges and
extreme weather events (El Mrini et al., 2012). The rapid develop-
ment of Tetouan coastline has already aroused the interest of
numerous studies, especially regarding its past and recent evolu-
tion in response to natural and anthropogenic forcing (Malek, 1995;
El Moutchou, 1995; Nachite et al., 2004; Niazi, 2007; Anfuso et al.,
2007; El Mrini, 2011; El Mrini et al., 2013). The short-term sedi-
mentary evolution of the shoreline have been analyzed by El Mrini
(2011), using digital elevation models obtained from three suc-
cessive surveys conducted during the stormy period of February
and March 2008. Results showed that seasonal beach changes are
not very significant; the most important variations were recorded
after storms. The type and mobility of beaches are function of their
curvature and distance from headlands, exposure to waves, grain
size and sediment supply. Morphological changes are faster and
more excessive in reflective beaches located north of the Cabo
Negro promontory; moreover, these beaches have a greater ten-
dency for erosion. Historical shoreline positions for Tetouan coast,
captured from multi-date aerial photographs for the period 1958 to
2003 (Niazi, 2007; Snoussi et al., 2008) and 1958e2007 (El Mrini,
 A. Satta et al. / Estuarine, Coastal and Shelf Science 175 (2016) 93e105 97

2011) showed an overall erosive trend of the shoreline. Eroded
beaches represent 70% of the coastline, while the accretion areas
account for only 14%. The overall coastline retreat is on average
80 m in the north and 45 m in the south coast. Average erosion rates
are 1.8 m/year in the northern coast (between M'diq and Fnideq)
and 1.0 m/yr in the southern sector (Cabo Ngro to Azla) (Niazi,
2007). In these conditions, and given the effects of current and
future human encroachment on the coast, local authorities are
faced with the increasingly complex task of balancing development
and managing coastal risks, especially coastal erosion and flooding,
and then be prepared to the unavoidable impacts of climate change.
In this prospect, the most valuable set of tools that may have direct
application in coastal development programs in both short and
long terms, and can assist the decision-makers in the imple-
mentation of preventive management strategies in the most sen-
sitive areas, is risk assessment. In this study, eight coastal units
were identified (Fig. 1 and Table 1) according to the different nat-
ural (river inlet, cape …) or anthropogenic structures (jetties …)
that influence the coastal sediment hydrodynamics processes. Each
unit is represented by one sandy beach.
hazards (H), which interacts with the vulnerability (V) and expo-
sure (E) factors. Definition of spatial attributes and selection of
variables is carried out on the basis of the relationships among
these three factors.
According to Fig. 3 (IPCC, 2014b), vulnerability and exposure are
influenced by Development (socio-economic pathways, adaptation
and mitigation actions and governance). Climate and Development
changes represent the key drivers of the different core components
(vulnerability, exposure, and hazards) that contribute to risk (IPCC,
2014b). In synthesis Risk can be considered as a function of three
factors:
Risk ¼ f (hazards, vulnerability, exposure)
As proposed by other indices in scientific literature (Davidson
and Lambert, 2001; Peduzzi et al., 2009) risk (R) follows a multi-
plicative formula, based on the standard formulation of risk “as the
product (or more correctly, convolution) of hazards, exposure, and
vulnerability” (Davidson and Lambert, 2001) as described in the
simplified equation:

3. Methodology
The Multi Scale Coastal Risk (MS-CRI) consists of an index-based
approach dealing with qualitative and quantitative spatial attri-
butes, representing physical, environmental and socio-economic
variables of the coastal system. The MS-CRI provides a simple nu-
merical basis for ranking sections of coastline in terms of their
potential for change that can be used by managers to identify re-
gions where risks may be relatively high and the results can be
displayed on maps.
The conceptual framework for risk and vulnerability makes
reference to the recently published WGII AR5 (IPCC, 2014a, 2014b),
which mainly focuses on risk. For this reason this paper focuses on
a coastal risk index instead of a coastal vulnerability index. The AR5
SPM (IPCC, 2014a) defines risk to climate-related impacts as the
result of the interaction of hazards with vulnerability and exposure
of human and natural systems (IPCC, 2014a). In the same report,
IPCC introduces the role of non-climate drivers (anthropogenic
climate change).
According to IPCC (2014b) Risk can be defined as “The potential
for consequences where something of human value (including humans
themselves) is at stake and where the outcome is uncertain. Risk is
often represented as probability of occurrence of hazardous events or
trends multiplied by the consequences if these events occur”. Fig. 3
shows Risk as a product of an interaction between hazards asso-
ciated with climate change and variability on one side, and the
vulnerability and its exposure to hazards on the other side (IPCC,
2014b). Climate and non-climate forcing acting on coastal haz-
ards, namely erosion and flooding, generate risk. In the MS-CRI
index forcing and hazard are incorporated in one factor called
Table 1
Coastal units adopted for the calculation of the risk index.
Coastal unit (from north to south) Length (m)a
Fnideq e Oued Negro 5540
Oued Negro e Marina Smir Port 4000
Marina Smir Port e Kabila Port 3450
Mdiq Port- Kabila Port 4200
Cabo Negro e Oued El Maleh inlet 3500
Oued El Maleh inlet e Oued Martil jetee 3000
Oued Martil jetee e Sidi Abdesslam 2750
Sidi Abdesslam e Cape Mazari 5000
a
From El Mrini (2011).
R¼H*V*E
Hence the magnitude of Risk depends on H and on the
component (V * E). The MS-CRI can thus be expressed as the
product of three factors: the Coastal Hazards (CH), the Coastal
Vulnerability (CV) and the Coastal Exposure (CE). The MS-CRI
equation can be written as follow:
MSCRI ¼ CH * CV * CE
These factors can be represented by multiple variables, keeping
in mind that the choice of good variables is crucial for the devel-
opment of a sound risk index (McLaughlin and Cooper, 2010). The
method integrates expert judgements and stakeholder preferences
“in order to aggregate quantitative and qualitative environmental and
socio-economic variables” (Torresan et al., 2012).
The distinguishing feature of the application of the MS-CRI at
the local scale (CRI-LS) lies in the definition of a methodology to
define the limits of the coastal hazard zone and the setback line.
The coastal hazard zone, intended as a coastal area where the risk
occurs, represents the spatial field of application of the method
(Satta, 2014).
The CRI-LS methodology is articulated in five main steps: 1)
Definition of the coastal hazard zone; 2) Variables choice and
ranking; 3) Assignation of weights to risk variables; 4) Aggregation
of variables, sub-indices and final index calculation; 5) Construc-
tion of the risk maps.
3.1. Definition of the coastal hazard zone
The CRI-LS methodology aims to integrate the provisions of the
Article 8 of the ICZM Protocol claiming the Mediterranean countries
to define a coastal hazard zone. In this paper, the coastal hazard
zone is defined as “the coastal zone affected by the occurrence of the
hazard effect, which has the potential to cause damage to, or loss of,
Beach name natural ecosystems, buildings, and infrastructure” (Satta, 2014). In
s Piedras
Tre other words, the coastal hazard zone is the landward limit of the
Restinga Smir buffer zone behind the coastline beyond which is defined as an
Kabila
acceptable level of risk produced by coastal forcing (Satta, 2014).
Golden Beach
Cabo Negro There is no specific definition in the scientific literature in terms of
Martil the best scientific methods to identify the coastal hazard zone
Sidi Abdessalam (Satta, 2014). Nevertheless, there are some operational definitions
Azla developed in the technical literature and related to the needs of
coastal planning. For a more detailed discussion on these
98 A. Satta et al. / Estuarine, Coastal and Shelf Science 175 (2016) 93e105
Fig. 3. Schematic of the interaction among the physical climate system, exposure, and vulnerability producing risk (Source: IPCC, 2014b).
definitions the reader is directed to Satta (2014). Taking into
consideration that the main hazards acting as forcing in the Med-
iterranean are erosion and flooding (Hemming et al., 2013) this
means that the overall hazard zone results from the overlaying of
the erosion hazard zone and the flooding hazard zone. As presented
by Niazi (2007) the maximum retreat of the coast of Tetouan, in the
worst SLR scenario for 2100, amounted to 77m, which means that
the erosion hazard zone of the study area is included in the 300m
side of the cell considered as a reference unit for this work. To
identify the hazard zone for flooding it is necessary to consider the
maximum water level (the inundation level) at the shoreline
resulting from extreme wave conditions (100-years Return period)
and extreme SLR. In Snoussi et al. (2010) the inundation levels for
the coastal area of Tetouan was assessed using empirical ap-
proaches. In particular, the inundation level was derived applying
the equation proposed by Hoozemans et al. (1993):
Dft ¼ MHW þ St þ Wf þPf
where Dft is the inundation level, MHW the mean high water level,
St the relative sea-level rise, Wf the height of storm waves and Pf
the sea-level rise, due to a lowering of the atmospheric pressure.
The maximum inundation levels was calculated by Snoussi et al.
(2008) using a mean value of 0.96 m for MHW, a storm wave height
of 6.20 m with a return period of 1 per 100 years, and a high estimate
0.86 m for sea-level rise. These values were related to the hydro-
graphic zero, which was 0.70 m below the general datum in
Morocco. The value of the maximum inundation level calculated by
Snoussi et al. (2008) is 7.64. The likely range of sea level rise in 2100
for the highest climate change scenario is 52e98 cm. However,
Anders Levermann, lead author of the sea level chapter for the IPCC's
Fifth Assessment Report notes that if sectors of the marine-based ice
sheets of Antarctic collapse, sea level could rise by an additional
several tenths of a meter during the 21st century.2 Thus, applying the
precautionary principle, as the upper value for SLR in 2100, it was
2
http://e360.yale.edu/content/feature.msp?id¼2698 (accessed March 25, 2015).
proposed a SLR of 1.5 m. By integrating the value of 1.5 m in the
equation, the result obtained is 8 m. To define the hazard zone for
flooding it was proposed a theoretical approximation. This work
analyzed the area under the value Dft considering a flood that
spreads inward without friction and is pushed inwards until the
height of 8m. This approximation is very precautionary since it is
evident that the friction of the ground would reduce the penetration
force of the flood that would stop long before reaching the altitude of
8m. This theoretical approximation was necessary to identify the
area in which the flooding could theoretically develop and on which
the risk index had to be applied.
This area, which represents the ‘coastal hazard zone’, is repre-
sented in Fig. 4.
3.2. Variables choice and ranking
For the calculation of the CRI-LS, was necessary to choose var-
iables describing the three sub-indexes: coastal hazard, coastal
vulnerability and coastal exposure. According to the classes pro-
posed by Satta (2014), with some adjustment adapted to the local
context of Tetouan, this paper proposed 6 variables for the Coastal
Hazards sub-index, 11 variables for the Coastal Vulnerability Sub-
Index and 2 variables for the Coastal Exposure Sub-index, for a
total of 19 variables. Variables and related rankings are presented in
Table 2.
The values for each variable, represented by a layer, are
normalized to a consistent ordinal or unit-less scale from lowest to
highest. The scaled layers are then averaged or added together to
come up with a score referring to its contribution to coastal risk.
According to Torresan et al. (2012), the allocation of scores assigned
to variables was performed using a 1e5 scale. The 1e5 scale, used
for every variable, standardizes the scoring system and enables
variables measured in different units to be combined mathemati-
cally (McLaughlin and Cooper, 2010). The maximum score 5 was
assigned to the highest risk class of the variable in terms of its
relative contribution to generate risk and in the same way the
minimum score 1 was assigned to the lowest risk class in the subset
of classes defined for each variable (Torresan et al., 2012).
 A. Satta et al. / Estuarine, Coastal and Shelf Science 175 (2016) 93e105
Fig. 4. Coastal Hazard zone for the region of Tetouan.
3.3. Assigning weights to risk variables
The integration of expert judgement is very important for the
allocation of scores to physical, natural and ecological variables and
the role of policy makers is fundamental in the evaluation of socio-
economic parameters (Satta, 2014).
With the aim to assign weights to each variable on a large and
participatory basis, a panel of scientific experts and local policy
makers was involved for the application of the CRI-LS to the coastal
zone of Tetouan. Each expert was asked to assign a score describing
the relative contribution of each variable to represent the sub-
indexes (i.e. hazards, vulnerability, exposure), taking into account
that the sum of the weights must be equal to 100%. The weighting
method proposed by Torresan et al. (2012) was chosen for this
paper. Variable judged to have higher influence on the final risk
were assigned a weight of 100%; on the other side, variables judged
to have no influence on the final risk were assigned a weight of 0%
(Torresan et al., 2012).
The values shown in Table 3 represent the average of the
opinions expressed by 12 between scientific experts and relevant
local stakeholders. For example, the variable LC (Land Cover) was
estimated to contribute 70% to the Coastal Exposure sub-index.
3.4. Aggregation of variables, sub-indices and final index
calculation
The numerical values for the three sub-indices (coastal hazard,
99
coastal vulnerability and coastal exposure) were weighted ac-
cording to value judgements and multiplied through a GIS raster
calculator function and the results were then categorized by
creating classes as a percentage of the maximum and minimum
possible scores. The formulas for deriving the three sub-indexes are
reported below:
Coastal Hazards (CH) sub-index ¼ [(WSLR * SSLR þ WSWH *
SSWH þ WMDP * SMDP þWDRO * SDRO þ WPGR * SPGR þ WTOUR * STOUR)
e 1] / 4}
Coastal Vulnerability (CV) sub-index ¼ {[(WLF * SLF þ WSLO *
SSLO þ WHSC * SHSC þ WELE * SELE þ WD * SD þ WRFR * SRFR þ WEH *
SEH þ WEDU * SEDU þ WP65 * SP65 þ WCPS * SCPS) e 1] / 4}
Coastal Exposure (CE) sub-index ¼ {(WLC * SLC þ WPDE * SPDE) / 5}
where S and W indicate respectively the score and the weight of the
variables of each sub-index.
The final CRI-LS index was computed by multiplying the three
sub-index values, as shown in the formula below:
CRI-LS ¼ CH sub-index * CV sub-index * CE sub-index
Each pixel score was calculated using a GIS raster calculator
function, which allows performing calculations on the basis of
existing pixel values and obtaining the results written into a new
100 A. Satta et al. / Estuarine, Coastal and Shelf Science 175 (2016) 93e105

Table 2
Choice and ranking of the variables used choice and ranking for the CRI-LS index.

Variable Description Score

1 2 3 4 5

Coastal hazards
Sea level rise (SLR) Level of the sea increased in one year. Satellite altimetry <1 1e1,6 1,7 - 2,4 2,5 - 3,2 n > 3,2
data provides accurate measures for a limited time range.
Storms (SWH) Average number of detected SWH above 95 <50 50e150 151e250 251e350 >350
percentile/year (SWHx95p), which represents the number
of events exceeding the long term (e.g. return period
Tr ¼ 100 yrs) 95 percentile of daily SWH.
Mean annual max Equal to the highest amount of precipitation received <16 16e32 33e64 65e128 >128
daily precipitation during the year, averaged over 30 year. Daily rainfall
(MDP) categories adapted from Alpert et al. (2002).
Droughts (DRO) Significantly driest winters experienced by the >36 36e12 11e12 13e36 < 36
Mediterranean regions during 1971e2010 relative to the
comparison period of 1902e2010. Low values indicate
a scarce sediment supply to beaches contributing
to erosion.
Population growth (PGR) Population growth (annual %) is the exponential rate of <0,1% 0,1% - 0,5% 0,51%e1% 1,01% - 2% >2%
growth of midyear population from year t-1 to t.
Tourism arrivals (TOUR) Number of tourists who travel to a country different < 0% 0% - 1% 1,01 - 5% 5,01% - 10% > 10%
from the one in which they are resident.
Coastal vulnerability
Landform (LF) Expresses the erodibility of the coastal zone. Scores are Hard Rock Soft Rock River deltas, Sandy shores Sandy shores
ranked according to the relative resistance of a shores shores estuaries and backed by and water plains
given landform to erosion. cobble beaches bedrock or
artificial
frontage
Coastal slope (SLO) Related to the relative risk of the shoreline retreat. > 0,1 0,1 - 0,05 0,049e0,034 0,033 - 0,02 < 0,02
Low sloping coastal regions should retreat faster.
Land roughness (LR) Represents the resistance to surface flow exerted by Urban Forest and Shrub land, Agriculture Bare areas
the land surface and is measured with the areas Water grasslands,
Manning's coefficient. bodies Sparse
vegetation
Historical Shoreline Percentage of eroded coast / Sediment budget. > 30% in 30% - 10% - 9,9% e -9,9% 10% -30% > 30% erosion
change (HSC) accretion in accretion stable erosion
Elevation (ELE) Represents the surface of selected coastal unit (pixel) 8 - 5,26 5,25 - 3,6 3,59 - 2,76 2,75 - 1 <1
within a specific class of elevation Xi
(e.g. 0.15m_Xi _ 0.3 m).
Distance from the Related to progression of the risk according to the > 4500 4500e2100 2099e900 899e300 < 300
shoreline (D) inland penetration of the flooding.
River flow regulation Represents the impact of dams on river flow no dams / Dams only in / Dams in the
(RFR) regulation in terms of sediment supply the minor largest tributary
(Oziurt, 2010). tributaries
Ecosystems health (EH) Expresses the contribution of the ecosystem as a No Slight signs Moderate Major Severe distortions
protection against storm surges, flooding and other detectable of distortions distortions with loss of
coastal hazards. change disturbance with loss of all species
50% of
species
Education level (EDU) Percentage of population whose level is equal at least > 60 60e44 43e28 27e10 < 10
to the level 3 of the international standard
classification of education (ISCED).
Age of population (P65) Percentage of population that is aged 65 years or older. <3 3e8,5 8,6e15 16e20 > 20
Coastal protection Percentage of the coastline with CPS (groynes, > 50 50 - 31 - 30e21 20e5 <5
structures (CPS) seawalls, revetments, etc.)
Coastal exposure
Land Cover (LC) The LC map from 2010 is a global land cover map Bare areas Shrub land, Forest and Agriculture Urban areas
at 300m spatial resolution. grasslands, Water bodies
Sparse
vegetation
Population density (PDE) The population density is derived by dividing the < 25 26e50 51e100 101e250 > 250
population count by the land area. It represents
population per Km2.

raster layer.
3.5. Construction of risk maps
The risk maps represent the main outcome of the CRI-LS
methodology developed in this paper. The values identified for
each variable are associated to a coastal spatial unit through a GIS
application. Each variable is represented by a set of polygons, each
of which is associated with a record, or a pair of values, one for the
class field and one for the score field (Satta, 2014).
Risk factors are calculated as the sum of the variables with their
relative weights for each significant coastal unit. The significant
coastal unit is an area of appreciable size, which makes the effects
of each variable meaningful. For this paper was considered a coastal
unit equal to a 300m  300m cell corresponding to the 300m
resolution of the Climate Change Initiative Land Cover of ESA3. The
3
CCI Land Cover - http://maps.elie.ucl.ac.be/CCI/viewer/index.php (accessed
February 2, 2016).
 A. Satta et al. / Estuarine, Coastal and Shelf Science 175 (2016) 93e105 101

Table 3
Data sources used for the variables and associated weights.

Variable Unit Data sources Weight (%)

Coastal hazards

Sea Level Rise (SLR) mm/y AVISO database 30%

Storms (SWH) Cm El Mrini (2011) & Niazi (2007) 25%
Mean Annual Max Daily Precipitation (MDP) mm/d http://www.water.gov.ma/ 10%
Droughts (DRO) Mm NOAA website 5%
Population growth (PGR) % http://www.hcp.ma/ 20%
Tourism arrivals (TOUR) % http://www.tourisme.gov.ma/ 10%
Coastal vulnerability
Landform (LF) Niazi, 2007 15%
Coastal slope (SLO) Niazi, 2007 10%
Land roughness (LR) Manning coefficient (n) associated to LC Chow, 2009 10%
Historical Shoreline change (HSC) % Niazi, 2007; 5%
El Mrini, 2011
Elevation (ELE) M GIS DEM 10%
Distance from the shoreline (D) M GIS 15%
River flow regulation (RFR) www.water.gov.ma 5%
Ecosystems health (EH) Ecological status by expert judgement Bibliography & expert judgement 10%
Education level (EDU) % www.hcp.ma 5%
Age of population (P65) % of pop. > 65 www.hcp.ma 5%
Coastal protection structures (CPS) % Expert evaluation 10%
Coastal exposure
Land Cover (LC) ESA map 70%
Population density (PDE) % of pop. / Km2 http://www.hcp.ma/ 30%

maps and the results of the analysis obtained with the application
of the CRI-LS to the case study of Tetouan are described and dis-
cussed in Section 4.
4. Results
The application of the CRI-LS to the coastal area of Tetouan led to
a ranking of the relative risk of the study area as well as the hazard
zone, in relation to potential coastal hazards generated and/or
exacerbated by climate and non-climate forcing. In particular, the
maps produced by CRI-LS allow identifying levels of risk not only
for the shoreline but also for the hinterland areas potentially
affected by flooding. The local risk map allows visualizing the
qualitative ranking of risk; to all variable data layers were assigned
a risk score and a weight. The CRI-LS equation, with weighted
variables, was applied to each cell of the coastal hazard zones for a
total of 1.284 cells. The maps obtained for the three sub-indexes are
shown in Fig. 5.
Hazards
The Hazards factor was constant throughout the area and equal
to 0,7125. The local scale defined for the case of Tetouan, does not
allow distinguishing the values between climate (SLR and SWH)
and non-climate variables (PGR and TOUR).
Vulnerability
To calculate the vulnerability values associated to each cell,
some simplifications were made to take into account the differ-
entiated effects of some variables on erosion and flooding. For
example the variable LF contributes just to coastal erosion hazard
while it has a neutral effect on flooding, so the scores of LF (from 1
to 5 according to the geomorphology), was assigned to all the cells
within line of 300 m of the ‘hazard zone’, while for the others a
constant value equal to 3 was assigned, as it does not affect the
impacts of coastal flooding. The scores of other variables, that
represent the susceptibility component of vulnerability (ELE, SLO,
D), decrease with the increasing distance from the coastline. Finally
the variable LR varies according to the Land Cover categories and
plays a key role in contributing to vulnerability, and in particular to
coastal flooding hazard.
Exposure
Most of the cells included in the ‘hazard zone’, are characterized by
urban land use, which is associated to a high population density (e.g.
Fnideq ¼ 1879,3 inhab/km2, Tetouan ¼ 4121,6 inhab/km2,
Martil ¼ 1164,5 inhab/km2). Apart from a few cells characterized by
land cover “bare areas”, and low-density housing (shown in green),
the whole area shows values of exposure from high to extremely high.
Risk
The CRI-LS scores were calculated by multiplying the cell values
of the three sub indices. The final risk rankings are dimensionless
numbers that judge the relative degree of risk of coastal zones to
coastal hazard, in relation to qualitative risk classes (i.e. extremely
high, high, medium, low, extremely low). Higher risk values imply
high risk in absolute terms and the proposed CRI-LS methodology
allows ranking the risk at the local scale of the studied area. The risk
scores obtained applying Raster Calculator range from 0.19 to 0.60.
The risk classes were obtained using a manual grouping method
with intervals of 0,2. The result is presented in Fig. 6.
The Local Risk map showed that cells characterized by
extremely high-risk (colored in red) are concentrated in the
waterfront. In addition to the higher risk of coastal flooding of the
waterfront, this is due to the cumulative effect of extremely high
exposure and vulnerability. The upper limit of the erosion risk zone,
even in the worst scenario SLR 2100, is less than 100m (Niazi,
2007). In order to better appreciate the differences, it would be
necessary to introduce a land cover with a higher resolution (e.g.
10m). Other authors assessed coastal vulnerability of Tetouan
littoral but with methods involving much less variables and
without taking into account the Coastal Hazard Zone. Niazi (2007)
assessed separately the physical vulnerability using the CVI pro-
posed by Gornitz et al. (1994), and the socioeconomic vulnerability
based on McLaughlin et al. (2002) method. The results of physical
vulnerability showed that 30% of the coast is classified at very high
risk (between Cabo Negro and Azla), 37% at high risk (between
102 A. Satta et al. / Estuarine, Coastal and Shelf Science 175 (2016) 93e105
Fig. 5. Maps of Coastal Hazards (A), Coastal Vulnerability (B) and Coastal Exposure (C)
M'diq and Fnidek) and 35% at moderate to low risk. Regarding the
socioeconomic vulnerability, based on variables such as population,
cultural heritage, land use and nature conservation, 17% (between
Martil and Sidi Abdessalam) exhibit very high vulnerability, 50%
have high vulnerability: namely the coastal urban and rural centers
of Fnidek, M'diq and Azla and regions with high tourism develop-
ment (Restinga and Cabo Negro). The remaining of the coast (33%)
has a moderate to low vulnerability. Anfuso et al, (2010) assessed
coastal vulnerability to erosion along the Northern part of the
Tetouan littoral, based only on the shoreline evolution (accretion/
retreat rate) and land use layers. They found that 10% and 27% of
investigated littoral fall under respectively very high and high
vulnerability as a result of combination of high erosion rates with
high capital land use. However, up to now no studies have evalu-
ated the coastal risk of this coast, using this number of variables.
One of the most relevant outcomes of the risk map with respect to
that of vulnerability map was that in the former even at distances
over 5km from the coast, the risk remains high, specially for the
following flood plains, considered from north to south:
- Restinga plain, located between Fnideq and M'diq is the nar-
rowest coastal lowland of Tetouan covering an area of 10 km2.
This little plain, shaped as a funnel is crossed by the Negro Wadi
and shows wetlands and salt marshes. The shore consists of a
sandy beach of the same name.
- Smir plain, located north of M'diq, covers an area of 12 km2 with
a general slope of less than 1.5% (SDAULTT 1993). The plain is
crossed by the Smir Wadi and ends in open water called merja
Smir that plays an important ecological role in the Mediterra-
nean region (Bayed and Scapini, 2005).
- Martil-Alila plain is the largest plain in the region with an area of
86 km2 and an elevation not exceeding 2m above sea level. The
plain is drained by the Mellah wadi in the north and Martil Wadi
to the south. Due to its low slope, drainage is difficult, which
makes it very exposed to floods. Inundated areas may last for
several months of the year.
The high levels of risk that characterize these areas, distant from
the coastline, are due on one hand to the low topography of the
flood plains and, on the other hand, to the high values of exposure.
5. Conclusions
The assessment of climate change impacts on the Mediterra-
nean coastal zones is driven by local priorities because the “vari-
ability of the coast, including human development patterns, result in
variable impacts and adjustments along the coast, with implications
for adaptation responses” (IPCC, 2007). In this sense, just through a
characterization of local features it is possible to assess the risk to
coastal hazards generated or exacerbated by climate and non-
climate forcing. This last aspect, together with the uncertainty of
SLR and marine storms projections, is one of the major potential
weaknesses for coastal risk assessment and for adaptation planning
in Mediterranean coastal regions. In fact, one of the most relevant
problems is the limitation on available data because in most loca-
tions, coastal data do not exist or have an uncertain quality due to
many other factors (e.g. calibration of the measuring devices) as
€
referred by different scientists (Snoussi et al., 2008; Ozyurt and
Ergin, 2010). The assessment of coastal risk is therefore the result
of a process of identifying, quantifying and ranking the variables of
all the components characterizing the coastal system at risk: social,
economic, environmental and political. In the specific case of this
research, the natural hazards by which risk was assessed are related
to the effects of climate and non-climate forcing. From the meth-
odological point of view, one of the first considerations that came
out from this analysis regards the relevance of the climate-forcing
factor, which we have called “coastal hazard” in assessing risk. At
the current level of knowledge variables like SLR and SWH are
measured and retrieved at the regional level reducing the accuracy
of the local data. In fact, available climatic variables used to assess
this factor were the same throughout the whole study area.
Therefore, the variables chosen for the calculation of the three
coastal risk sub-indexes are highly dependent on data availability at
the local level. It would be theoretically possible to have more in-
formation taking into account the non-climatic variables such as
PGR and TOUR that can vary locally in a very significant way. In this
case, by assigning larger weights to the non-climatic variables, it
would be possible to highlight the differences, even if substantial,
 A. Satta et al. / Estuarine, Coastal and Shelf Science 175 (2016) 93e105
Fig. 6. Tetouan coastal zone risk map
between the different municipalities of the same region. Particu-
larly relevant in the analysis of the local risk is the contribution of
the Exposure factor that can be further refined with a higher
definition of Land Cover (e.g. 10m  10m). In this sense, Tetouan is
highly urbanized with a high-density housing. The built up area in
the coastal plain reached 95% in 2007 (Nachite, 2009), which makes
the area extremely exposed to coastal risk. In terms of decision
support, the application of the Multi-Scale Coastal Risk Index at the
local scale, CRI-LS, identified the areas at risk within the hazard
zone defined with the maximum inundation level under the worst
SLR Scenario in 2100. Given the speed of development that the
study area is facing and the likely impacts of more extreme whether
events related to climate change, the CRI-LS can prove to be the an
appropriate tool to support coastal managers in developing adap-
tation measures and in allocating resources to respond to coastal
erosion and coastal flooding to the areas at highest risk.
Coastal managers need accessible and user-friendly tools to
103
define risk scenarios and to select sound planning and manage-
ment strategies. A dynamic and iterative ICZM process should
integrate the risk-assessment method for setting coastal manage-
ment priorities. In this sense the CRI-LS explores hazard, vulnera-
bility and exposure to describe the coastal system (Satta, 2014) and
provides a framework for making informed decisions.
The main advantages of CRI-LS in supporting the ICZM process
can be summarized as follows: 1) integrates non-climate drivers
(e.g. tourism pressure) and interactions between climate and non-
climate drivers (e.g. socioeconomic indicators); 2) proposes a local
spatial scale (300m resolution), which is compliant with the nature
of the hazard to be assessed; 3) provides a method to identify the
setback line as defined in the ICZM Protocol4; 4) provides a user
4
ICZM Protocol - http://eur-lex.europa.eu/legal-content/EN/TXT/HTML/?
uri¼CELEX:22009A0204(01)&from¼EN (accessed February 2, 2016).
104 A. Satta et al. / Estuarine, Coastal and Shelf Science 175 (2016) 93e105
friendly and reliable method to explore the characteristics of the
studied coastal ecosystem in terms hazard, vulnerability and
exposure; 5) variable aggregation and quantification of risks using a
linear equation do not require high scientific expertise making this
method applicable also in less developed coastal societies.
Furthermore the application of the MS-CRI in the ClimVar
project showed its applicability to very diversified geographical
contexts. Nevertheless, like for any index-based tool, there are
some limitations to its application, including subjectivity regarding
variable selection and weighting, problems with aggregation to
different scales and difficulties validating the results. To overcome
these limitations, there is a need to strengthen the technical and
scientific capacity of coastal decision makers to apply these
methodologies, including the improvement of coastal monitoring
and datasets construction. Moreover, testing and applying the CRI-
LS in other study areas will reduce the subjectivity in the selection
of variables. It is necessary to make existing the CRI-LS widely
available and fill the gaps in coastal manager needs. It is also
necessary to incorporate into education and training for key pro-
fessions, including engineering, architecture, planners, parks
managers and local government. The capacity of the CRI-LS to
measure risk and to produce risk maps, even in contexts of data
scarcity and resources constrains, can significantly contribute to the
preparation of coastal hazards adaptation strategies. To check
consistency and to confirm the potential of this method for repli-
cation or generalization, additional work at other coastal areas
around the Mediterranean is required for comparative purposes.
Acknowledgments
This work was carried out as part of the ‘‘Strengthening the
Knowledge Base on Regional Climate Variability and Change”
project prepared by Acclimatise and its associates. This project was
funded by the United Nations Environment Programme (UNEP)
under the GEF PROJECT ID 3990 in the framework of the ClimVar
project and implemented by Plan Bleu, Regional Activity Centre of
the UNEP-MAP.
References
Alpert, P., Ben-Gai, T., Baharad, A., Benjamini, Y., Yekutieli, D., Colacino, M.,
Diodato, L., Ramis, C., Homar, V., Romero, R., Michaelides, S., Manes, A., 2002.
The paradoxical increase of Mediterranean extreme daily rainfall in spite of
decrease in total values. Geophys. Res. Lett. 29, 1e31. http://dx.doi.org/10.1029/
2001GL013554.
Anfuso, G., Martinez del Pozo, A., Nachite, D., Benavente, J., Macias, A., 2007.
Morphological characteristics and medium-term evolution of the beaches be-
tween Ceuta and Cabo Negro (Morocco). Environ. Geol. 52, 933e946.
Anfuso, G., Martinez del Pozo, J.A., Nachite, D., 2010. Coastal vulnerability in the
Mediterranean sector between Fnideq and M’diq (North of Morocco). Comptes
rendus de l’Acade mie bulgare des Sciences, Ge ographie physique.
Geomorphologie 63 (4), 561e570.
Bayed, A., Scapini, F., 2005. Impact des ame nagements hydrauliques, touristiques et
urbains sur l'equilibre et le fonctionnement de la lagune et des marais de Smir.
Ecosyste mes co ^ tiers sensibles de la Me diterranee : cas du littoral de Smir.
Travaux de l'Institut Scientifique, Rabat 4, pp. 75e85.
Benavente, J., Bello, E., Anfuso, G., Nachite, D., Macias, A., 2007. Sobreelevacio n
debida a temporales y cambios producidos en las playas del litoral Marroquí.
Revista Cuaternario y Geomorfología 21, 13e25.
Davidson, R., Lambert, K., 2001. Comparing the Hurricane disaster risk of U.S. coastal
counties. Nat. Hazards Rev 3 (132), 132e142. http://dx.doi.org/10.1061/(ASCE)
1527-6988(2001)2.
El Moutchou, B., 1995. Dynamique co ^ tie
re actuelle et e
volution morphose dimen-
taire de la frange littorale me diterrane enne entre M'diq et Oued Laou (Re gion
de Tetouan, Maroc nord-occidental). Fac. des Sciences. Rabat, Univ. Mohammed
V. These de 3e me cycle: 131 p.
El Mrini, A., 2011. Evolution morphodynamique et impact des ame nagements sur le
littoral tetounais entre Ras Mazari et Fnideq (Maroc nord occidental).
PhD.thesis. University Abdelmalek Essaadi, Morocco and Universite  de Nantes,
France.
El Mrini, A., Anthony, E.J., Maanan, M., Taaouati, M., Nachite, D., 2012. Beach-dune
degradation in a Mediterranean context of strong development pressures, and
the missing integrated management perspective. Ocean & Coast. Manag. 69,
299e306, 2012.
El Mrini, A., Anthony, E.J., Taaouati, M., Nachite, D., Maanan, M., 2013. A note on
contrasting morphodynamics of two beach systems with different backshores,
Tetouan coast, northwest Morocco: the role of grain size and human-altered
dune morphology. J. Coast. Res. (Special Issue No. 65), 2013.
ETC-CCA, 2011. Methods for assessing coastal vulnerability to climate change. Eu-
ropean Topic Centre on Climate Change Impacts. Vulnerability and Adaptation -
Technical Paper 1/2011.
Gornitz, V.M., Daniels, R.C., White, T.W., Birdwell, K.R., 1994. The development of a
coastal risk assessment database: Vulnerability to sea-level rise in the U.S.
southeast. J. Coast. Res. (Special Issue No. 12)
Hemming, D., Agnew, M.D., Goodess, C.M., Giannakopoulos, C., Ben Salem, S.,
Bindi, M., Nejmeddine Bradai, M., Congedi, L., Dibari, C., El-Askary, H., El-
Fadel, M., El-Raey, M., Ferrise, R., Grünzweig, J.M., Harzallah, A., Hattour, A.,
Hatzaki, M., Kanas, D., Lionello, P., McCarthy, M., Aranda, C.M., Oweis, T.,
Sierra, J.P., Psiloglou, B., Reale, M., S
anchez-Arcilla, A., Senouci, M., Tanzarella, A.,
2013. Future Climate Projections. Chapter 3. In: Navarra, Antonio,
Tubiana, Laurence (Eds.), Case studies. Regional Assessment of Climate Change
in the Mediterranean, vol. 3.
Hoozemans, F.M.J., Marchand, M., Pennekamp, H.A., 1993. Sea Level Rise. A Global
Vulnerability Assessment. Delft Hydraulics, Delft/Rijkswaterstaat, The Hague,
184 pp.
lede
tection et du SIG  tude de l’e
volution
Ibrahimi, I., 2003. Application de la te a l’e
morphologique et du transit se dimentaire du littoral de Te touan. Doctorat,
Univ. Mohammed V, Faculte  des Sciences, Rabat, 180p.
IPCC, 2007. Summary for Policymakers. In: Parry, M.L., Canziani, O.F., Palutikof, J.P.,
Van der Linden, P.J., Hanson, E.E. (Eds.), Climate Change 2007: Impacts, Adap-
tation and Vulnerability. Contribution of Working Group II to the Fourth
Assessment Report of the Intergovernmental Panel on Climate Change. Cam-
bridge University Press, Cambridge, UK, pp. 7e22, 2007.
IPCC, 2014a. Summary for policymakers. In: Field, C.B., Barros, V.R., Dokken, D.J.,
Mach, K.J., Mastrandrea, M.D., Bilir, T.E., Chatterjee, M., Ebi, K.L., Estrada, Y.O.,
Genova, R.C., Girma, B., Kissel, E.S., Levy, A.N., MacCracken, S., Mastrandrea, P.R.,
White, L.L. (Eds.), Climate Change 2014: Impacts, Adaptation, and Vulnerability.
Part A: Global and Sectoral Aspects. Contribution of Working Group II to the
Fifth Assessment Report of the Intergovernmental Panel on Climate Change.
Cambridge University Press, Cambridge, United Kingdom and New York, NY,
USA, pp. 1e32.
IPCC, 2014b. In: Field, C.B., Barros, V.R., Dokken, D.J., Mach, K.J., Mastrandrea, M.D.,
Bilir, T.E., Chatterjee, M., Ebi, K.L., Estrada, Y.O., Genova, R.C., Girma, B.,
Kissel, E.S., Levy, A.N., MacCracken, S., Mastrandrea, P.R., White, L.L. (Eds.),
Climate Change 2014: Impacts, Adaptation, and Vulnerability. Part A: Global
and Sectoral Aspects. Contribution of Working Group II to the Fifth Assessment
Report of the Intergovernmental Panel on Climate Change. Cambridge Univer-
sity Press, Cambridge, United Kingdom and New York, NY, USA, 1132 pp.
L.P.E.E, 1997. Estimation de l’e tat morphologique des plages. Laboratoire Publique
des Essais et d’Etude. Rapport de la Direction des e tudes Techniques. Ministe res
des travaux publics, 450 p.
Malek, F., 1995. Evolution morpho-se dimentaire de la frange littorale
Me diterrane enne au Nord de Te touan entre M’diq et Fnidek (Maroc Nord
occidental): Interfe rence des causes naturelles et anthropiques et impacts des
ame nagements portuaires. Doctorat de 3e me cycle. Universite  Med V, Faculte 
des sciences, Rabat, 103p.
McLaughlin, S., Cooper, J.A.G., 2010. A multi-scale coastal vulnerability index: A tool
for coastal managers? Environ. Hazards 9 (3), 233e248.
McLaughlin, S., McKenna, J., Cooper, J.A.G., 2002. Socio-economic data in coastal
vulnerability indices: constraints and opportunities. J. Coast. Res. (Special Issue
No. 36)
Mcleod, E., Poulter, B., Hinkel, J., Reyes, E., Slam, R., 2010. Sea-level rise impact
models and environmental conservation: a review of models and their appli-
cation. Ocean & Coast. Manag. 53, 507e517, 2010.
Nachite, D., 2009. Le de veloppement touristique du littoral de la re gion Tanger-
Tetouan: une e volution vers des sce narios non de sirables ? In: Domínguez
Bella, S., Maate, A. (Eds.), Geología y Geoturismo en la Orilla Sur Del Estrecho De
Gibraltar. MCN - UCA Cadiz, ISBN 978-84-9828-224-5, pp. 59e78. ISBN.
Nachite, D., El Moutchou, B., Anfuso, G., Benavente, J., Bello, E., Macias, A., 2004.
Morphologie et e volution recente du littoral entre Fnideq et M’diq (Te touan, NE
du Maroc). Geogaceta 35, 43e46.
Niazi, S., 2007. Evaluation des impacts des changements climatiques et de
le
l’e vation du niveau de la mer sur le littoral de Te touan (Me diterranee occi-
dentale du Maroc): Vulne rabilite
 et Adaptation (Phd thesis).
€
Ozyurt G, 2007. Vulnerability of Coastal Areas to Sea Level Rise: a Case of Study on
Go€ksu Delta. Thesis submitted to the Graduate School of Natural and Applied
Sciences of Middle-East Technical University. January 2007.
€
Ozyurt, G., Ergin, A., 2010. Improving coastal vulnerability assessments to sea-level
rise: a new indicator-based methodology for decision makers. J. Coast. Res. 26
(2), 265e273.
Peduzzi, P., Dao, H., Herold, C., Mouton, F., 2009. Assessing Global Exposure and
Vulnerability Towards Natural Hazards: the Disaster Risk Index.
Satta, A., 2014. An Index-based Method to Assess Vulnerabilities and Risks of
Mediterranean Coastal Zones to Multiple Hazards. Doctoral dissertation sub-
mitted in October 2014 to the Department of Economics Ca’Foscari University of
Venice, Italy.
Satta, A., Venturini, S., Puddu, M., Firth, J., Lafitte, A., 2015. Application of a Multi-
 A. Satta et al. / Estuarine, Coastal and Shelf Science 175 (2016) 93e105
scale Coastal Risk Index at Regional and Local Scale in the Mediterranean. PLAN
BLEU Technical Report - September 2015.
Snoussi, M., Ouchani, T., Niazi, S., 2008. Vulnerability Assessment of the Impact of
Sea-level Rise and Flooding on the Moroccan Coast: the Case of the Mediter-
ranean Eastern Zone.
Snoussi, M., Niazi, S., Khouakhi, A., Raji, O., 2010. Climate change and sea-level rise:
a GIS-based vulnerability and impact assessment, the case of the Moroccan
coast. In: Maanan, M., Robin, M. (Eds.), Geomatic Solutions for Coastal Envi-
ronments Book. Nova Publishers.
View publication stats
105
Torresan, S., Zabeo, A., Rizzi, J., Critto, A., Pizzol, L., Giove, S., Marcomini, A., 2010.
Risk assessment and decision support tools for the integrated evaluation of
climate change on coastal zones. In: Swayne, D.A., Yang, Wanhong, Voinov, A.A.,
Rizzoli, A., Filatova, T. (Eds.), Proceedings of the International Congress on
Environmental Modelling and Software Modelling for Environment's Sake, Fifth
Biennial Meeting, Ottawa, Canada, 2010.
Torresan, S., Critto, A., Rizzi, J., Marcomini, A., 2012. Assessment of coastal vulner-
ability to climate change hazards at the regional scale: the case study of the
North Adriatic Sea. Nat. Hazards Earth Syst. Sci. 12, 2347e2368.