*2.1. Methods*

The proposed CVI considers the following 10 variables:


In addition to the six variables described by [13], four new variables have been proposed: Emerged beach width; dune width; width of vegetation behind the beach and Posidonia oceanica. The new proposed variables, representative of the Mediterranean coast, allow us to evaluate the ability of "natural systems" to dissipate the wave energy. In particular, sandy beach-dune systems constitute the natural barrier protecting coastal areas against flooding due to storm surge and wave impacts. Furthermore, the effects of a well vegetated beach and seagrass Posidonia oceanica on wave energy have significant implications for coastal protection.

All variables have been divided into three typological groups: Geological, Physical process and Vegetation.

The Geological variables are:


The Physical process variables are:


The Vegetation variables are:


A stretch of coast is divided into a number of transects (or cross-sectional profiles of the beach) in order to assess its vulnerability. Each transect is characterized by a control area 0.5 km wide. Variables are ranked on a linear scale from 1–5 in order of increasing vulnerability.

The CVI is obtained by the square root of the product of the vulnerability scores assigned to each variable divided by the total number of variables:

$$\text{CVI} = \sqrt{(\mathbf{a} \cdot \mathbf{b} \cdot \mathbf{c} \cdot \mathbf{d} \cdot \mathbf{e} \cdot \mathbf{f} \cdot \mathbf{g} \cdot \mathbf{h} \cdot \mathbf{i} \, \text{l})/10} \tag{1}$$

where a = Geomorphology, b = Coastal slope, c = Shoreline erosion/accretion rates, d = Emerged beach width, e = Dune width, f = Relative sea-level change, g = Mean significant wave height, h = Mean tide range, i = Width of vegetation behind the beach, l = Posidonia oceanica.

CVI values are classified in four different categories (low vulnerability, moderate vulnerability, high vulnerability and very-high vulnerability) using percentiles as limits.

#### *2.2. Data*

In the following, the data sources used to define the 10 variables are listed.


Table 1 shows the range of vulnerability for the 10 variables. Regarding the variable geomorphology, the ranges of vulnerability considered are those proposed by [65]. The ranges of vulnerability for the variables of relative sea-level change, mean significant wave height and shoreline accretion/erosion rate have been chosen in agreement with those proposed by [37]. Regarding the variable coastal slope, the values have been chosen considering previous studies carried out for the Mediterranean coast [4,36,37]. In particular, the range chosen are those proposed by [4]. Regarding the variables emerged beach width, dune width and width of vegetation behind the beach, the ranges of vulnerability have been defined in consideration of the characteristics of the Italian and Mediterranean area [66]. Furthermore, the available data (regional orthophotos) made it possible to verify the similarity of these considered values with those typical of the Mediterranean environment. Finally, for the variable mean tide range, the ranges of vulnerability have been chosen in accordance with those proposed by [37] but the scores (linear scale from 1–5) are different. This assumption, in agreement with [13], is based on the concept that, in general, microtidal (tide range < 2.0 m) and macrotidal (tide range > 4.0 m) are characterized by high and low risk, respectively. The reasoning is based primarily on the potential influence of storms on coastal evolution, and their impact relative to the tide range. For example, on a tidal coast-line, there is only a 50 percent chance of a storm occurring at high tide. Thus, for a region with a 4.0 m tide range, a storm having a 3 m surge height is still up to 1 m below the elevation of high tide for half a tidal cycle. A microtidal coastline, on the other hand, is essentially always "near" high tide and therefore always at the greatest risk of inundation from storms [13]. Mediterranean area is a microtidal environment and the coast of Apulia has a tide range < 1 m. As such, the range of vulnerability, as mentioned above, are those proposed by [37] but the assigned scores are the inverse.

Other researchers (e.g., [12,67]) claimed the opposite; the large tidal range coast-lines were assigned a high-risk classification, and microtidal coasts received a low risk rating. The reasoning for this is that although a large tidal range dissipates wave energy, limiting beach or cliff erosion to a brief period of high tide, it also delineates a broad zone of intertidal area that will be most susceptible to inundation following long-term sea-level rise. Furthermore, the velocity of tidal currents depends partially on the tidal range. High tidal range is associated with stronger tidal currents that are capable of eroding and transporting sediment [67].


**Table 1.** Ranges of vulnerability for the considered variables.

#### **3. The Study Area**

The proposed CVI index has been applied to a stretch of the coast of the Apulia Region, Southern Italy, between the marinas of Torre Canne and Villanova (Figure 1).

**Figure 1.** Case study area—Marinas of Torre Canne and Villanova, Apulia Region, Southern Italy.

Starting from the north, Torre Canne (Figure 2), the first stretch of about 7 km (up to Torre San Leonardo), which corresponds to the first 15 to 24 investigated transects, is characterized by beautiful beaches, which the Apulia Region has intended to protect by establishing with Regional Law No 31/2006, the Coastal Dunes Park. The park covers about 1.000 hectares. In the protected area there are many priority habitats, strongly threatened for their intrinsic fragility and for being located in areas at risk, but also habitats of the Community interest representative of the biogeographical reality of the Community territory. In the area there are beaches, consolidated dunes, retrodunal ponds and fossil dunes. The remaining 5 km of coastline, ranging from Torre San Leonardo to the port of Ostuni marina (Villanova) which make up the remaining nine transects of the study area, are jagged cliffs and consist of a series of coves with small beaches surrounded by Mediterranean vegetation.

**Figure 2.** Case study area and related transects.

The study area is mainly devoted to seaside tourism, and there are indeed many tourist accommodation facilities (hotels, resorts, etc.).

#### **4. Results**

A Geography Information System (GIS) platform has been used to better process the data. The stretch of coast has been divided into 24 transects; geomorphology includes very-high vulnerability and moderate vulnerability;coastal slope values are <2% varying between 0.75% (min) and 1.52% (max), while shoreline erosion/accretion rates is classified as very-low to very-high vulnerability. The value of the relative sea-level change is constant at low vulnerability and mean tide range is constant at very-high vulnerability. Mean significant wave height is constant at high vulnerability. Emerged beach width includes high vulnerability and moderate vulnerability, while dune width includes very-high vulnerability, high vulnerability and moderate vulnerability. Width of vegetation is classified as very-low to very-high vulnerability. Finally, Posidonia oceanica predominantly shows a low vulnerability since it is present in many transects.

The estimated minimum CVI value calculated for the case study is 30, while the maximum value is 300. The CVI mean is 123.40, the median is 84.85. The classes of CVI values have been divided into "low vulnerability" (green), "moderate vulnerability" (yellow), "high vulnerability" (orange) and "very-high vulnerability" (red) categories, respectively, on the basis of 25th, 50th, and 75th percentiles [13] as summarized in Table 2.


**Table 2.** Vulnerability categories.

Table 2 shows the vulnerability categories, while Table 3 shows the vulnerability value associated to each variable and the estimated CVI values for each transect (a = Geomorphology, b = Coastal slope, c = Shoreline erosion/accretion rates, d = Emerged beach width, e = Dune width, f = Relative sea-level change, g = Mean significant wave height, h = Mean tide range, i = Width of vegetation behind the beach, l = Posidonia oceanica).


**Table 3.** Vulnerability value associated to each variable and CVI values for each transect.

Figure 3 shows a screenshot of the GIS page with the CVI values for the case study area.

**Figure 3.** CVI value for each transect.

#### **5. Discussion**

For the case study area, the most important variables are geomorphology, shoreline erosion and accretion rates, beach width, dune width, width of vegetation behind the beach and Posidonia oceanica, since the other variables are constant. As described above, the variable geomorphology mainly includes sandy beaches (very-high vulnerability) and low cliffs (moderate vulnerability), while shoreline erosion and accretion rates attain values between low vulnerability and moderate vulnerability.

The variable emerged beach width attains values between moderate vulnerability and very high vulnerability, as the beaches are not very large but rather narrow, while for the variable dune width in the area, the dune is present only in some transects characterized by no significant widths. Width of vegetation behind the beach is classified as very-low to very-high vulnerability while Posidonia oceanica is present in many transects.

In particular, transects from 1 to 3 are characterized by a sandy beach, with a low coastal slope and moderate emerged beach width; it should be noted that there is the absence of dune and vegetation, with constructions built close to the shoreline; for transect 1 and 3 the vulnerability is partly mitigated by the presence of Posidonia oceanica.

Transects 4 and 5 present a moderate vulnerability due to the presence of a modest dune, vegetation and Posidonia oceanica.

In transect 6 and 7, the value of vulnerability increases in relation to the absence of Posidonia oceanica, and for transect 6 a greater erosion is observed.

The transects from 8 to 13 present vulnerability that is predominantly moderate in relation, especially due to the presence of vegetation and Posidonia oceanica.

The transects from 14 to 17 (Torre San Leonardo), are characterized by the transition from sandy beach with dunes to low cliffs; in this stretch of coast there is an increase in vulnerability due to the absence of Posidonia oceanica and to the considerable reduction of the dune and the vegetation; this stretch is characterized also by an intensive land use with important population centers.

The transects from 18 to 24 are characterized by low and moderate vulnerability for the presence of Posidonia oceanica and vegetation.

It is important to highlight that in index-based methodologies, such as CVI, the availability of reliable and up-to-date databases is crucial. Variables like geomorphology and coastal slope can be considered stable since present negligible changes in time, while for the relative sea level change, mean tide range, and mean significant wave height, consolidated, international databases exist. For the variable shoreline erosion/accretion rates, reliable and up-to-date databases may not always be available [68]. For this variable, as well as for emerged beach width, dune width and width of vegetation behind the beach, it may be necessary to consider field measurements or the use of aerofotogrammetry. The field measurements are more precise but require significant investment and are limited in time and space. While the use of aerofotogrammetry is less accurate, it can be extended to larger areas. A more recent and useful tool for creating databases on these variables is the multispectral processing of images from satellites; these images present higher resolution with pixel processing and gradation based on photographic interpretation procedures (multispectral processing). This allows activating procedures for semi-automatic and/or automatic recognition of spatial elements. In fact, the increasing availability, resolution and spatial coverage of satellite imagery in recent years now provides a powerful alternative to derive reliable, global scale shoreline data. In this direction, in many recent studies the satellite images coupled with image processing techniques have been used [69–73].

#### **6. Comparison between Two Index-Base Methods**

In the present paper, the proposed CVI has been compared with the index-based method proposed by [37], postponing its verification to a later study by more complex process-based models (e.g., [74,75]).

The objective of the comparison is to evaluate, compared to an index similar in structure and range of vulnerability for each variable, the effects of further variables not yet taken into account. Specifically, it should be noted that the use of additional variables such as emerged beach width, dune

width, width of vegetation behind the beach and posidonia oceanica, is based on the consideration that these variables can be useful to better characterize the Mediterranean coasts, especially the low-lying coastal areas.

Regarding the Coastal Sensitivity Index (CSI) proposed by [37], it uses the following physical variables: geomorphology, coastal slope, relative sea-level rise rate, shoreline erosion or accretion rate, mean tidal range and mean wave. This index was applied to the southern coast of the Gulf of Corinth, Greece. The obtained results are summarized in Table 4 and shown in Figure 4.


**Table 4.** Comparison between the proposed CVI and the CSI.

**Figure 4.** Proposed CVI and CSI values.

Generally, the two investigated methods show similar results. Some differences have been found likely due to the four variables proposed in the CVI to evaluate the ability of "natural systems" to dissipate the wave energy: emerged beach width; dune width; width of vegetation behind the beach and Posidonia oceanica.

In particular, for the cases of transects 9–13, the presence of Posidonia oceanica and the width of the vegetation determine a reduction of vulnerability in relation to the values obtained with the CSI. However, in the case of transects 16 and 17, the absence of Posidonia oceanica and dunes, the narrow beaches and the small width of the vegetation, determine an increase in vulnerability.

#### **7. Conclusions**

The CVI is a useful method for the assessment of the relative physical vulnerability of a stretch of coastline to the effects of climate change. The present paper proposes a CVI formulation, suitable for the Mediterranean coasts, that considers 10 variables and allows us to evaluate the vulnerability with respect to SLR, storm surges and waves action. In the following, the main conclusions of the study are as follows.

The tailored index CVI indicates that the dune width and the geomorphology are the most important drivers in building a regional index in terms of increasing the risk of flooding in this region. Regarding dune width, different transects are characterized by the absence of dunes or small-width dunes. Therefore, the relative vulnerability scores are mostly 5 (very-high vulnerability) and 4 (high vulnerability). For the geomorphology variable, most of the case study area consists of sandy beaches with a relative vulnerability score of 5 (very-high vulnerability).

On the contrary, width of vegetation behind the beach, shoreline erosion/accretion rates and Posidonia oceanica variables show a negligible influence. Width of vegetation behind the beach and shoreline erosion/accretion rates are classified as very-low to very-high vulnerability, while Posidonia oceanica is present in many transects.

The application of the proposed index shows the feasibility of the index and the possibility of using the CVI to make assessments on coastal vulnerability with respect to climate change.

The aim of the future research is to validate the proposed index by comparing it with the more complex numerical models in order to make the index a useful tool for coastal planning and management.

**Author Contributions:** D.P. Data curation, Formal analysis, Methodology, Validation, Writing-original draft, Writing-review & editing; F.D. Writing-review & editing; L.R. Validation (collaboration in the CVI application), Writing-original draft (collaboration in paper writing); F.P. Validation (collaboration in the CVI application); G.R.T. Conceptualization, Supervision, Writing-review & editing.

**Funding:** This work was funded by the Apulia Region (Italy), through the Regional Cluster Project "Eco-Smart Breakwater", Grant #S6LU5I7.

**Conflicts of Interest:** The authors declare no conflict of interest.

#### **References**


© 2018 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).
