CADSEALAND

This paper present a critical summary of main results from the Work Package 4 of the CADSEALAND project. Some open questions, uncertainties and needs for future work are also discussed.

The area under study, located in the western side of the Northern Adriatic sea, is the coast of the Emilia-Romagna region, Italy. This coastal area, flat and sandy, extends from the Po river delta north to the south border of the Emilia –Romagna region, and has a total shore length of about 130 km. A large part of this coastal area suffers, such as most of the Adriatic shore, strong coastal erosion. Factors controlling the coastal retreat are a complex interactions between sediment supply, subsidence, meteo-climatic changes and the strong human impact occurred in the last 50 years.
Because the Emilia Romagna coastal zone is a very important area from economic and social point of view, to contrast coastal retreat the Region government approved the guidelines for integrated coastal zone management - ICZM .(delibera consiliare n.645 del 20 gennaio 2005). The guidelines indicates the main fields of activity and the type of actions that should be realized from the various institutions and stakeholders. Moreover the guidelines provide some indications about the administrative tools and type of project that should provide the resources to realize the actions. The INTERREG project CADSEALAND is one of this tools.

The Integrated Informative System to Support Coastal Protection Strategies

During the CADSEALAND project the concept and main parts of an Integrated Informative System to Support Coastal Protection Strategies were developed in the Emilia Romagna region.
The main parts of the system are the Coast Information System, the meteo-marine data base and the mesoscale numerical models providing hindcast and forecast of the sea waves, sea currents and sea level in the Adriatic sea. The system products are accessible via an user interface allowing to extract data, GIS products and the numerical models output.
The system is characterised by a distributed architecture, the data bases, numerical models and other components which are installed on several computers belonging to ARPA-SIM and GSSS. A large number of applications can be developed starting from the core system now available. Applications can be grouped into two main categories: the applications dedicated to support the ICZM by the Regional Government and local institutions and the applications dedicated to the support of the coastal hazard early warning by the national and regional civil protection services. The first type make use mainly of the hindcast results and of the data over long time periods, the second type mainly uses forecasts.
Some of the output and data are available on public WEB pages, such as the SIM page (www.arpa.emr.it/sim/) in the section “MARE”, where the sea state and oceanographic forecasted maps are daily available. In the web site of the Environmental Agency, ARPA (www.arpa.emr.it/mare/) several kind of information related to the coastal system are grouped, including the model output and meteorological forecast. Some of the outputs related to the physical characteristics of the coastal system are currently available in the GSSS Web page in the thematic area dedicated to the coast. The GSSS web is a portal useful to switch several interactive cartography. http://www.regione.emilia-romagna.it/wcm/geologia/canali/costa.htm.

Applications during the CADSEALAND project

Some applications of the integrated system for coastal zone management purposes were realized during the CADSEALAND project, at different time scales:

1. the climatological time scale: the general long term evolution of the Emilia-Romagna coast during the last 100 years and detailed coastal dynamic of the area between Rimini and Cesenatico were analysed. The evolution of Rimini-Cesenatico coastal sector was compared with the assessment of the storminess in the Adriatic sea during the last 50 years.
2. The short time scale: the analysis of measured wave data and simulations over a 5 year time period was performed to identify the occurred sea storms and their seasonal distribution and trend.
3. Some case studies were analysed using all the information made available by the informative system. The focus of these studies was to assess the role of the meteorological and oceanographic factors in determining the impact of a strong sea storm on a beach and dune system.

The results for each time scale are summarized below:

Results for the climatological time scale (100 and 50 years)

The Emilia-Romagna coastal evolution at 100s years time scale (1811-1998) shows three areas with different behaviour:

• A Northern area (Po di Goro) that shows a fast accretion of the distributary channels of the modern Po Delta;
• Central area (Ravenna) is characterized by a strong differentiation among delta fronts and beaches between the river mouths (“intermouth”); the first ones were progressively eroded through the last century while intermouths consequently accreted in the same period. As result the coast progressively become more straight. A Southern area (Rimini-Cesenatico) that remained through time almost stable with shoreline variations of a few hundreds meters that tend to be strongly reduced after the II World War due to the intensive man made intervention in the form of coastal protection structures.

These long-term coastal dynamics of northern and central Emilia-Romagna coast are strictly related the capacity of rivers to develop their delta fronts and are likely related to changes in the sediment supply rate of the Po and Apennine rivers. The evolution of the southern area, having historically a quite stable coast-line behaviour, was instead strongly affected by human impact. However the absence of detailed meteorological and wave data covering 100 years time interval does not allow to understand the contribution of meteo-marine events to such shoreline variations.
Coastal dynamics at the 50 years (1943 - 2005) time scale was studied in detail in the ”Rimini Cesenatico Pilot Area”, which is characterized by a great development of hard costal defences and coastal urbanization. The study demonstrated that such a strong human intervention is the most important factor affecting coastal evolution. Between 1970-1980 the area records a peak in construction of coastal defences, producing a reduction of shoreline retreat in some critical areas. In the same period even regional laws supported an integrated coastal defence strategy by the reduction of river bed excavations.
The analysis of storminess trends in the Northern Adriatic area does not show strong variations in the 50s years (1951 ¿ 2004) time interval, even if a minor decrease in the average storm frequency has been recorded after late seventies, and it might have slightly contributed to beach stabilization.
Geometrical data on shoreline changes indicates that parallel coastal defences have not the same efficacy across the whole coastal sector. They generate beach accretion only at the river mouths, where the sand supply is guaranteed by river floods, while their positive effect progressively decreases northwards. Moving away from beaches at river mouths they do not produce accretion any longer and the only role they have for coastal protection is to generate shoreline stabilization.
The asymmetry of coastal accretion at river mouths indicates that northward littoral drift prevails at all temporal scales.

Results for short time scale (5 years)

There is a general lack of wave data on the Emilia-Romagna coastal area for the period 2000-2004, therefore the wave forecasts by the WAM wave model were used to calculate the significant wave height and direction during that period.
The model was previously verified by means of scores that compare model outputs versus all the wave data available for the Adriatic Sea and made available from the National Buoy Network (RON). Verification against data recorded by the closest observation point available with a significant statistical number of data, the Ancona Buoy, show that:

• The wave height is well described by the model (the bias is less than -0.05 m), although the forecast values are quite flattened compared to observations (RMS ~ 0.5 m)
• The main direction of southern waves is well accurately forecasted, but with a slight shifting toward east (the bias on wave direction range between -5 and -25 degrees)

The bias and RMS values are referred to the forecast range [+24 , +48 hours] in order to avoid the model spin-up effect. Considering the encouraging results of the validation the forecast data were used for the isolation and evaluation of historical sea storm.
A sea storm extreme events was considered when a significant wave height higher than 1.5 m for a minimum duration of 6 hours was forecasted or observed. This definition of sea storm was adopted after a review of the relevant literature. The 5 years (2000 - 2004) time series of the wave height on three points (Ravenna, Cesenatico, Misano) near shore (9 m depth) and offshore (30 m depth) were extracted and processed in order to calculate the seasonal frequency of occurrence of sea storm along the entire Emilia-Romagna coast.
Results indicate that sea storms mostly occur during autumn and winter and with higher strength (in terms of significant wave height) than the rest of the year. On the Emilia Romagna coastal area the main direction from which waves come during storms is the North-East in winter and East in autumn.
Sea storm frequency of occurrence has grown during the winter period of the last three years considered, whereas it has been more or less constant for the autumn period. The maximum number of storms occurred during the 2003-2004 winter period.
The comparison between the sea storm frequency and direction computed for the Emilia Romagna coastal area by the WAM model and the same variables observed by the Ancona and Punta della Maestra buoys suggests that the “wave climate” is different in the three locations, both for the wave height and direction.
These results confirm that there is a strong demand to measure the waves parameters along the Emilia Romagna coast.

Case studies (Lido di Dante, September - November 2004)

During September and November 2004 strong sea storms caused hazards for open sea and harbour activities, damage to seaside tourist facilities and severe erosion to the Emilia-Romagna littoral.
The work on the case study site of Lido di Dante has quantified the impact of the exceptional storm, with combined high waves and storm surge, that happened on 24-26 September 2004. It was decided to focus the analysis on one of the few remaining areas in the Region with coastal dunes.
These storms were analysed in the framework to the CASEALAND project to test and verify the models and to start to study the link between the meteo-marine conditions and the evolution of coastal morphology. The full set of data coming from observations and numerical models, made available from the coastal informative system, was employed to describe the sea storm and its impact on the beach.
The analysis of meteo marine data indicate two main storm events: the September 24 storm and the November 13-14 storm.
During the September 24 storm a maximum surge toward the Emilia¿Romagna coastline was described at 4:00 pm UTC combined with a fast north-east “Bora” wind episode.
During the November 13-14 storm a strong wind set up of the sea level toward the northern Adriatic coast (south-east “Scirocco” wind) was described at 09 UTC of November 13. The day after (09 UTC November 14) a wind rotation from south-east to north-east caused a strong wind-sea.
Both the episodes can be considered as an example of a typical, although not frequent, storm surge combined with a severe storm.
The storm run-up levels eroded and damaged the dune ridge, at sites where this was at elevation lower than 4 m above mean sea level. It can be concluded that future dune reconstruction should consider the elevation of wave run-up during storms and associated surges.
The study of coastline variability in the same area has found that the erosive crises started to become evident during the winter of 2003/2004, the year with the maximum number of storms, as pointed out by the 5 years analysis. However, a limited stability and capacity of the profile to recover after storms was still present. It should be noted that the storms of the autumn and winter 2004/2005 found a beach profile weakened by the storms occurred during the previous winter season (2003/2004). The beach had only a limited amount sand stock available, and this extremely high event generated at many point irreversible damage to beach/dune system. Direct measurements, obtained combining the LIDAR dataset with topographic surveys undertaken before and after a small beach replenishment by the local authorities, have proved that the positive effects of the protection scheme can hardly be seen, unless a consistent amount of sediment is inside the nearshore budget. The replenishment was undertaken at the dune base and quickly eroded. The sand was partially lost offshore and partially remained on the intertidal area. A larger replenishment volume, in part distributed over the upper beach and along the dune front could have had more positive effects.

Concluding remarks, open questions and needs for future work

The present work demonstrate that most of the basic and fundamentals tools needed to assess, forecast and hindcast the meteorological and marine factors acting on the Emilia-Romagna coast are now available. On the other hand the Coastal Informative System (SIC) provide many of the geomorphological data needed to describe and model the coastal area, especially to assess the erosion trends at almost any time-scale. Moreover the system can be applied to support the risk assessment and management of the coastal area.
The exercises of applying the forecasting system to study the coastal dynamic and its links with the meteorological and sea conditions, evidenced many uncertainties and needs to improve and expand the system itself. The modular design of the system considers these needs. The open architecture adopted allows a continuous improvement and makes easy to replace some of the components of the system.
The study case of Lido di Dante, used to assess the impact of an exceptional storm, remarked that both waves and sea levels (e.g. storm surges) must be modelled, in order to predict the effects of sea storm on the coastal zone. The present version of the system can provide the full set of information needed for early warnings and/or archived output for fast environmental assessment. These information are the forecast and hindcast of wind, wave and sea level.
On the other hand, to estimate the impact of the storm on the beach and verify the models, a monitoring platform placed in the nearshore must be available. The monitoring platform should measure directional waves and sea level. These measurements should be coupled with surveys to quantify variations of the shore profile before and after a high energy event. Despite the fact that many sporadic nearshore surveys have been undertaken in the area and temporary physical measurements were carried out, there is a lack of real or near-real time wave data for the Emilia-Romagna coastal area. Similarly there is a scarce knowledge of the impact of storm events.
For coastal management purposes there is a need to estimate the probability of occurrence of the high energy events. The answer to this question can be only partially obtained from the 5 years analysis of wave storms where it was calculated that the number of sea storms ranges between 9 and 13 per year. That analysis did not consider the combined occurrence of wave storm and storm surge, that seems to be the critical issue to determine a strong impact on the beach. In order to obtain an exhaustive assessment of the probability of occurrence of that kind of high energy events a reanalysis over a long time period should be performed running meteorological, wave and ocean models. The reanalysis could be a main task for future work.
The application of the system to the total balance analysis of the entire beach surface , shows a reduction of the shoreline retreat after hard coastal defence erection in critical areas . The slight decrease of average storm frequency after the late seventies, could have contribute to beach stabilization even if a strong correlation does not appear.
The coastal dynamics analysis points out that there is a strong interaction between coastal defence structures and sediment transport, especially in the nearshore zone. That kind of interaction occurs at a spatial scale not described by the current version of the oceanographic and wave models. On the other hand the present set of numerical models can provide boundary conditions to run more detailed models. To study the small scale processes a very high resolution coupled wave and currents model should be run. Finally this could be matched to process-based models, to simulate the beach profile evolution during exceptional events. However, unless such a model will be calibrated with specific field data, its predictive capability will remain unknown.