**Piero Ruol \*, Luca Martinelli and Chiara Favaretto**

ICEA Department, Padua University, v. Ognissanti 39, 35129 Padova, Italy; luca.martinelli@unipd.it (L.M.); chiara.favaretto@dicea.unipd.it (C.F.)

**\*** Correspondence: piero.ruol@unipd.it; Tel.: +39-049-827-7905

Received: 30 May 2018; Accepted: 24 July 2018; Published: 26 July 2018

**Abstract:** This paper discusses the key aspects of the recent Coastal Plan of the Veneto Region (IT). Its aim is to propose a single mitigation strategy for coastal erosion that is valid for the whole Veneto Region, and possibly elsewhere, as well as a method to assign a priority level to any action. The suggested mitigation action against erosion depends on urbanization level, beach width, as well as cross-shore and long-shore sediment transport. The criterion used to give a priority level to mitigation actions is based on a vulnerability index that takes into account erosive tendency, existing coastal flooding hazards, coast value, environmental relevance, tourist pressure, urbanization level, the presence of production activities, and cultural heritage. A sample case featuring the littoral of Rosolina is also provided and includes a site description, the sediment budget, critical issues and possible mitigation measures.

**Keywords:** coastal plan; erosion, coastal flooding; sediment budget; mitigation strategies guidelines; littoral cell

### **1. Introduction**

In the recent past, in Italy, when coastal managers at the Regional Authorities were faced with coastal erosion problems, they were frequently guided by an empirical approach based on simple transport equations, or even solely by their intuition. More recently, however, the perspective has changed somewhat and it has become evident that the complexity of coastal systems needs to be studied through a homogenous and multidisciplinary approach with reference to large spatial and temporal scales that take into account a wide range of phenomena and topics. Coastal zones must be analysed from several points of view (e.g., geophysical, biological, socioeconomic, political, cultural, historical) and with different approaches (e.g., research, planning, operational purposes). Successful management requires a thorough understanding of the physical processes impacting the coast to create a strategic vision of the future, establishing a framework to guide future actions [1–3]. In light of this, local authorities are expected to produce a medium-term coastal plan that can effectively reduce and manage the risks that natural processes (e.g., storms, floods and erosion) pose to human health, the environment, cultural heritage, and businesses. They are also required to produce a flooding vulnerability map that includes coastal inundation in compliance with EU Directive (2007/60/EC). As a result, many of the available studies (e.g., Hinkel et al. [4], Weisse et al. [5], Toimil et al. [6]) that include the most recent IPCC Assessment [7] show that European coasts are exposed to erosion, rising sea-levels and climate change and discuss how to manage this threat at large scale.

Guidance for shoreline management has been provided by several recent research projects. The following studies comprise a demonstration site on the North-Eastern coast of Italy, and were carried out in collaboration with local stakeholders: RITMARE Flagship Project (Antonioli et al. [8], Bezzi et al. [9]), CAMP Italy Project, COASTANCE (Montanari and Marasmi [10]), DELOS, (Zanuttigh et al., [11]) COASTGAP-MED, MEDSANDCOAST-ENPI, COASTAL Mapping-DG

MARE, EUROSION [12], THESEUS (Zanuttigh [13]), MICORE (Ciavola et al. [14]) or RISC-KIT (Armaroli et al. [15]). Specific open-source tools and approaches have been developed to support decision-making processes (Zanuttigh et al. [16], Torresan et al. [17]; Vafeidis et al. [18], van Dongeren et al. [19]; Stelljes at al. [20]).

In order to implement the monitoring and coastal plans, that can be adopted by the local authorities, and/or in order to draw the vulnerability maps required by the Flood Directive, the valuable results obtained by tools such as the THESEUS DSS require an engineering synthesis, e.g., studies like those described in Preti et al. [21], Petrillo et al. [22], Ruol et al. [23]. Such studies provide simple guidelines for the design of erosion and coastal flooding mitigation measures that follow coherent and rational criteria.

This paper hence aims to propose erosion mitigation criteria that are valid for the whole Veneto Region, and possibly for adjoining coasts. It also aims to provide practical guidelines on how to interpret coastal monitoring analysis, select when, where and what mitigation measures should be adopted, and suggest a methodology for assigning a priority level to any action.

The analysis here presented originates from the practical need to investigate the Venetian littoral and to establish a coastal plan based on a balanced combination of scientific rigorous approach and expert, discussion-based, assumptions. The Northern Adriatic coast is subject to rapidly evolving pressures from a range of drivers, including natural and anthropogenic ones (e.g., rapid morphological evolution of Po River Delta [24], human-induced subsidence caused by fluid withdrawal [25], changing wave climate [26]), all of which require an integrated approach. The study also outlines a detailed monitoring plan and shows that the mitigation measures adopted depend on the monitoring results.

In addition to this introduction, this paper includes two main sections and a concluding paragraph. First, the analysis methods are described, which include a set of available mitigation options (whose effectiveness is based on the behaviour of previous works on the Venetian littoral) and the rational criteria for their selection. The method is applied to the whole Venetian littoral, together with an example of the mitigation measures adopted on a short stretch of coast. Lastly, conclusions are drawn.

#### **2. Methodology**

The geography of the Venetian littoral is presented in Section 3.1. The Coastal Plan of the Veneto Region (IT) is divided into a descriptive phase and a design phase.

The descriptive phase involves collecting all of the available coastal data, possibly integrated or obtained with numerical tools. These data may then be used to extend knowledge and reduce uncertainties. The coastal plan includes a sediment budget assessment that provides essential information for the subsequent steps.

The design phase carries out a critical interpretation of the sediment budget results, selects the most appropriate mitigation option based on a univocal criterion and prioritizes mitigation actions. All major issues along the littoral zone were discussed with a wide stakeholder group.

The core of the method (common to both the descriptive and design phase) involves subdividing the regional coast into littoral cells and organising all of the information and results into a single geographical information system (Coastal GIS). The concept of "sediment cells" [12,27] allows for a better understanding of sediment transport patterns. These cells are stretches of coast with similar characteristics bordered by morphological features, such as river mouths, inlets and port dams, meaning that, in the absence of major obstacles to long-shore currents, sediment is relatively free to move inside the morphological feature.

In addition, each cell is divided in half to form two semi-cells (S-C). Dividing these cells into two half (not necessarily equal in length) has practical advantages for the sediment budget balance, as the boundary conditions between them are morphologically continuous. Therefore, the long-shore sediment transport is continuous here and easy-to-compute by using wave climate and local beach

orientation, and a new set of independent equations is added, allowing a more reliable assessment of the entire budget.

The Coastal GIS is a large database that is easy to access and update, and could be shared with all the stakeholders and local managers. The use of GIS for coastal management has expanded rapidly during the past decade (Bartlett and Smith [28], Wright and Bartlett [29]), and is suited to the Williams [30] approach, based on "getting", "reordering" and "refining" the information. The collected information is suitable to be integrated within a CoastalME type framework [31].

Figure 1 shows the general structure of the proposed methodology, whereby the information gathered from both the descriptive and design phases is stored in the Coastal GIS.

**Figure 1.** Breakdown of coastal plan activities. The Coastal GIS includes both data (measured or computed) gathered in the descriptive phase and the results following the design phase.

#### *2.1. Descriptive Phase*

#### 2.1.1. Data Collection and Harmonisation

Data and available measurements are collected, harmonised and stored in the Coastal GIS. Data for sediment budget analysis consist of all the variables that characterise local coastal morphology (i.e., shorelines, bathymetries, DTMs, grain size distribution, dune characterisation, subsidence, etc.), the forcing (i.e., wave, wind, tide, surge, currents, etc.) together with existing defence measures, and the sources and sinks of sediments (i.e., fluvial sediment transport, littoral sediment transport, a detailed history of past nourishments and dredgings).

Other information is required to define the constraints (e.g., areas with special legal protection or regulation, urban planning), and the value of the area in general (i.e., environmental relevance, urban and tourist pressure, local economy, cultural heritage, etc.). Additional data may be relevant to conducting flood-risk and vulnerability assessment (e.g., inland mapping, land use, emergency plans).

#### 2.1.2. Data Integration by Numerical Modelling

Numerical models are frequently used in coastal planning, since they are useful for integrating any available information, especially on forcing, coastal flooding and sediment transport, e.g., [13,21–23,27,31].

Models such as ECMWF, AdCirc, Wavewatch III help provide a detailed description of the wind, currents and wave climate [32], both in terms of extremes and average values. In some cases, it is useful to statistically analyse the forcing provided by extensive databases (e.g., NOAA's Historical Hurricane Tracks [33]).

Evaluation of the coastal flooding risk is a specific task that may be integrated using raster images included in the Coastal GIS as described in [34]. The flooding probability derived by the model is stored in form of vulnerability maps [35].

After the data collection phase, other numerical models (e.g., Mike21, X-Beach, LitPack, Gencade) are used to achieve a complete and homogeneous description of the sediment transport discharge for each of the sediment cells identified, taking into account the grain size distribution. The simulations evaluate the river sediment supply and long-shore and cross-shore sediment transport at regional level during the predefined time interval selected for the sediment balance. These data are essential for the assessment of the sediment budget and hence for the design phase.

#### 2.1.3. Assessment of Sediment Budget

The sediment budget is essentially a mass balance equation applied to a specified time interval. It is convenient to subdivide the coastline into a number of small stretches and apply the balance to each stretch *i*:

$$
\varepsilon\_{\bar{i}} = \partial V\_{\bar{i}} - Q\_{\bar{i}} \Delta t = 0,\tag{1}
$$

where *∂Vi* is the volume of accretion or erosion estimated by a comparison of the bathymetries surveyed at the beginning and at the end of the time range Δ*t*, and *Qi* is the net sediment discharge shown in Figure 2 and given by:

$$Q\_{\dot{i}} = -Q\_{LS(North)\dot{i}} + Q\_{LS(South)\dot{i}} + Q\_{\dot{r}i} - Q\_{CR\dot{i}} + Q\_{N\dot{i}} - Q\_{Di} \tag{2}$$

where *QLS* is the long-shore sediment transport (*North* identifies the discharge at the northern boundary and *South* at the southern boundary in an ideal beach aligned in the North-South direction); *QF* is the additional river sediment supply; *QCR* is the cross-shore sediment transport; and *QN* and *QD* are the volumes added or subtracted due to nourishment or dredging respectively. Subsidence and sea level rise do not affect the sediment balance directly but they have the same effect as generalized erosion as they alter the marine accommodation space. In order to preserve the equilibrium profile, the accommodation space must be balanced by beach nourishment.

**Figure 2.** Sediment balance diagram. Littoral cells are limited by morphological features (continuous line). The inclusion of river sediment transport in the balance equations would be straightforward if the cell control volume extended to the dash-dot box.

The balance in Equation (1) forms a system of 2*n* − 1 equations, where each stretch *i* corresponds to a semi-cell (S-C). As anticipated at the beginning of Section 2, boundary conditions between the two halves of the same littoral cell are morphologically continuous, whereas the littoral cell boundaries are placed in correspondence with morphological elements, e.g., river mouths, that can complicate the assessment of the long-shore sediment transport.

River mouths need to be considered with special attention. The inclusion of river sediment transport in the balance equations would be straightforward if the control volume extended to the two S-Cs adjacent to the mouth (i.e., the dash-dot box in Figure 2) since river supply is equivalent to nourishment. However, Equation (1) is applied to each S-C. In the adopted scheme, river sediment transport is divided and allocated to both the adjacent S-Cs, and the fine fraction losses are treated as additional off-shore sediment transport. The approach is similar to the one described by Samaras and Koutitas [36,37]. To provide a single procedure for all of the cell boundaries, the long-shore transport *QLS* is calculated even where the boundary is a groin, a port, a river mouth, etc. However, in presence of the latter, further calculations are required to establish *QF*. *QF* is given by the total river supply (Q*RIVER* in Figure 2) minus the fine fraction *QCR* that will be transferred seaward, minus the previously calculated value of *QLS*. Obviously, in the trivial case of a stable river mouth, the river supply is equal to the long-shore sediment transport, and the resulting value *QF* is zero. It is also immediately clear that an insufficient river supply results in a negative value of *QF*.

The system in Equation (1) is coupled since long-shore sediment transport represents a mixed term, e.g., the *QLS(South)* of the Northern S-C is equal to the *QLS(North)* of the Southern one (dotted arrow in the middle of the cell *i* showed in Figure 2). It is solved using a compensation of error technique based on a matrix of uncertainties given a priori. Error compensation by least squares adjustment is obtained by solving an overdetermined system of equations based on the principle of least squares of observation residuals. It is used extensively in the disciplines of surveying, geodesy, and photogrammetry [38]. Guidelines on the evaluation of the a priori uncertainties of each term can be found, for instance, in [39]. Support of the stakeholders is essential for this phase.

#### *2.2. Design Phase*

#### 2.2.1. Mitigation Options and Selection Criteria

Initially, based on the descriptive phase, the specific causes inducing erosion should be found, e.g., reduction of river sediment transport, increased subsidence rate, etc. Appropriate action should then be geared towards reducing these causes, e.g., providing sediment bypass in the presence of river dams and limiting extractions from the soil, etc. Similarly, reintegration of damaged structures/environmental areas shall be considered, with dune restoration being a typical measure that provides a reserve of sand in the event of storms as well as a safety against coastal flooding.

Engineering solutions for mitigation of flood and erosion risks described in [39] include both active methods, based on the reduction of the incident wave energy, such as the use of wave energy converters, floating breakwaters and artificial reefs, and passive methods, consisting of increase in overtopping resistance of dikes, improvement of resilience of breakwaters against failures, and the use of beach nourishment (possibly with innovative layout [40]) as well as tailored dredging operations. Suggestions on design optimization, optimal placement, and efficiency from an ecological perspective are outlined in [41].

Non-structural mitigation options are discussed in [42] where it is pointed out that they should be considered as part of a potential portfolio in which their combination transcends their sum, rather than standalone measures. Also, mobilization of stakeholders in the implementation process is an important issue to achieve effective results.

This sub-section puts forward the more common options and discuss their suitability for the coast of the Veneto region, which is mainly low and sandy.

Low sandy beaches can be subdivided into two categories: (1) "urbanized coasts", which are intensely developed and have a high economic value, for which the main goal is to defend urban and tourist activities, possibly preserving a large beach; (2) "non-urbanized coasts", natural littoral, for which the aim is to preserve the environmental value.

The proposed mitigation options follow this classification.

**1. Urbanized low sandy coasts**: The criterion is based on two main physical characteristics: the net long-shore sediment transport (*net LST*) and the gross long-shore sediment transport (*gross LST*). The rate of cross-shore sediment transport (*CST*) is also taken into account when *LST* values are low. Figure 3 shows a scheme of the guidelines for this type of coast. In all the cases, nourishment is the main strategy adopted (e.g., [43,44]), possibly together with structures (e.g., submerged barrier, low-crested structure, groins) or/and other measures (e.g., down-drift maintenance, up-drift bypass).

**Figure 3.** Guidelines for erosion mitigation strategies (*div(LST)* = divergence of long-shore sediment transport, *CST* = cross-shore sediment transport) for urbanized low sandy coasts.

In particular, for very low *LST* (i.e., both *net* and *gross*), obstacles to *LST* are considered fairly ineffective at intercepting sediments. When the *CST* is also negligible (case "c" in Figure 3), causes of erosion are independent of coastal dynamics, and plain nourishment is the most suitable action [45]. However, when *CST* is appreciable (case "b" in Figure 3), the building of a submerged barrier is suggested. This structure is placed at the breaking line, deeply submerged. The expected piling up [46] and wave transmission [47] are small. It is considered effective in stabilising the cross-section profile, since breaking always occurs at the same point, and therefore the bar is stabilised, minimising the offshore *CST* associated with offshore bar migration [48]. For very large *CST* (case "a" in Figure 3), a more effective solution is required. The candidate is the low crested structure (*LCS*) scheme with small gaps [11], which essentially confines the sediment in the protected area, i.e., the area bordered by the shoreline, the lateral low crested groins and the offshore barrier. Note that a large piling up is expected to occur in this type of protected area, forming an obstacle to coastal dynamics and possibly resulting in some down-drift maintenance being needed [49]. The same solution is appropriate even when the *gross LST* is large but the *net LST* is small.

For large values of *net* and *gross LST*, the best solution is probably a battery of groins, as it tends to stabilise the shoreline in a saw-tooth shape and reduces *LST*, with benefits for the protected area [50]. However, the downdrift area (at the end of the battery of groins) must be properly maintained, and a bypass system may help reduce the sediment resource needed. In the event of extreme *net LST* (which only occurs with extreme *gross LST*), permeable groins may be considered. By reducing, but not blocking, the littoral current, the velocity differential between the velocity seaward and in the pile-groin fields is smaller than with impervious groins [51]. The primary objective of the design is to reduce the littoral current velocity to an extent that rip currents and large-scale circulations in the groin field are minimised.

When *net LST* is appreciable but *gross LST* extreme, groins are evidently not very effective [52] and do not compensate the minor *CST* caused by the increased wave reflection as it interacts with the structure. Plain nourishment is, therefore, the solution suggested.

In areas where the shoreline is expected to move significantly, such as in proximity of a river mouth, for which exceptional riverine sediment transport may occur, fixed structures should not be built (they may become unsuited to the modified shoreline).

**2. Non-urbanized low sandy coasts:** The criterion for natural beaches (Figure 4) is based on two physical characteristics, i.e., *net LST* and beach width. According to Van der Nat et al. [53], who classify the degree to which coastal designs are nature-based using criteria for ecosystem-based management, the "building with nature" approach is particularly suited to this category. Three soft stabilization measures are taken into account: two measures envisage supplying sand to the beach in the form of distributed or concentrated nourishment, and one contemplates increasing sand-dune volume.

When the beach is wide and *net LST* large, i.e., sediment transport direction is well-defined, concentrated nourishment updrift is the least impacting solution for reducing erosion in this location, as it limits the impact of the works on a single spot and relies on littoral dynamics to redistribute sand downdrift (an extreme example is the "Sand Engine" project in the Netherlands [54]). For small *net LST*, the ability to redistribute nourishment is limited, so nourishment should cover the whole of the eroding coastline. Wherever the beach is narrow (i.e., may be completely eroded within a short time period), and a drastic modification of the natural environment is not viable, sand may be accumulated creating new dune systems, thus reinforcing the existing beach [55,56].

**Figure 4.** Guidelines for erosion mitigation strategies (*div (LST)* = divergence of long-shore sediment transport) for non-urbanized low sandy coasts.

All of the aforementioned mitigation options require some degree of nourishment, making resource availability a critical issue. It is, therefore, necessary that the plan includes the location of any sediment resource along the coast, offshore and/or inland. In most cases, the sediment balance is very helpful for finding the position and size of any coastal stocks [44]. Clearly, this sediment stock makes for excellent nourishment since it belongs to the same environment and is the most economical to derive. Some care must be taken when planning the annual volumes to be dredged from these areas, and a permanent monitoring plan is necessary.

The aforementioned measures mitigate the risk of erosion. In the event of flood risk, however, strategies must be integrated with other measures [57]. Dune restoration may be considered the first candidate in order to defend the inland from marine inundation [16]. Seawalls, that induce a local erosion due to reflection, are cost-effective solutions to limit the overtopping, especially when the parapet is appropriately shaped [58,59]. *LCS* and other detached parallel structure, by reducing the wave energy incident directly to the coast, may have a significant impact in the reduction of run-up. Similarly, large nourishments have a very beneficial effect, both favouring energy dissipation, due to a reduced water depth, and reducing run-up, due to milder foreshore slope.

In addition to these measures, other anti-erosion and anti-flooding actions are available for both urbanized and natural coasts: managed realignment; urban planning; stabilization of river mouth (embankments to confine river discharge, plus bypass or sediment stock for updrift nourishment); lagoon dredging for environmental purposes; revetments (only where the unprotected area has no environmental and tourist interests); beach dewatering systems [60,61]; artificial reefs for habitat restoration; wave energy converter farms acting as coastal defence [62,63], seasonal interventions (e.g., sand accumulation, sand bags). It is also recommended that innovative solutions be lab-tested to ensure that they are effective in cost/benefit terms before they can be safely used on-site [16].
