Erosion hazard assessment along the Capitolo coast (Monopoli, southern Italy)

Abstract

Erosion affects about 25% of the coast of the Apulia region, with the severest consequences being along sandy coastlines. Beach retreat is mainly due to the decrease of solid material transport, the destruction of dune deposits and the building of docks that obstruct the longshore current. A survey was conducted around Capitolo, the main sandy coastline near Bari, to examine the hazards affecting beaches and to provide guidelines for the management of human activities along the coasts. The survey was carried out along a coastal stretch of about 3 km, extending in a South Easterly direction from 17° 21′ 30″ E, 40° 54′ 20″ N to 17° 23′ E, 40° 53′ 40″ N. The map, at a scale of 1:3000, shows three levels of hazard, determined as follows: (i) the state of preservation of the dunes and watercourses, based on the geomorphological map of the Apulia region; (ii) data collected from July 1989 to April 2008 by the Monopoli buoy of the National Wave Measuring Network; (iii) the shoreline change rates computed within a geographic information system (GIS). A matrix was established taking into account of these features, weighting each parameter in order to develop a hazard-level measurement for shoreline lengths of about 25 meters. The results show that all Capitolo beaches have a high level of shore preservation hazard. The results of this study should be a factor in any decisions made on the management of the present day village, both existing commercial activities and the future development of the area.

Keywords:

• coast
• erosion
• hazard
• GIS
• Apulia

1. Introduction

The study presented in this paper is part of a more comprehensive survey carried out along the whole Monopoli coastline which aims to identify sites where developments such as tourist resorts can be established with minimal impact on the natural evolution of the shoreline. The survey focused on the beach of Capitolo, which is the only sandy area in the rocky Monopoli coast.

The area covered by the study is on the Adriatic (eastern) side of the Murge plateau, about 50 km SE of Bari, in the municipality of Monopoli. Capitolo has a long stretch of sandy beach, which is heavily developed and given over to private lidos. The beaches are the main attractions for visitors to the central part of the Apulia and dozens of cottages and hotels have been built near the coastline, destroying dunes. Building development has taken place over a considerable period of time without taking into account the coast's natural dynamics, under which human activities cause a great deal of damage.

A detailed survey of the features influencing beach dynamics was conducted in order to suggest to local authorities the best coastal management policy for the area. Three levels of hazard are shown on the Main Map, based on changes in the coastline (1994–2010), sea weather events (wave height measured from the Monopoli buoy) and a geomorphological survey showing dune preservation and ‘lame mouths’.

2. Geological and geomorphological setting

The Murge is a karstic area which represents the central part of the Southern Apennines foreland (Apulian Foreland), formed by the thick carbonate Mesozoic sequence known as ‘Calcari delle Murge’ Group. It is overlaid by thin calcareous and marly-calcareous deposits of Tertiary and Lower Pleistocene known as the ‘Calcareniti di Gravina’ Formation. The landscape all around the Murge plateau is characterized by a number of marine terraces produced by the superimposition of regional uplift and glacio-eustatic sea-level changes that have occurred since the Middle Pleistocene (Ciaranfi, Pieri, & Ricchetti, 1988; Mastronuzzi & Sansò, 2002a). The area exhibits a drainage-network formed by regularly spaced, incised valleys draining toward the coastal area. These channels are locally named ‘lame’ (Gioia, Sabato, Spalluto, & Tropeano, 2011). According to Mastronuzzi and Sansò (2002a), the valleys originated cyclically by sapping processes along the coasts of the uplifting Murge during periods of high sea-level.

The Monopoli coastline is formed, overall, by a series of headlands and inlets, interrupted by pocket beaches where ‘lame’ reach the sea (Andriani & Walsh, 2007). The transport of bedload in ‘lame’ occurs only during extreme storm events; as a result, for most of the year the valleys are dry. The southernmost part of the Monopoli coast consists of about 3 km of beaches in the vicinity of Capitolo. Along this stretch of shore, mid-Holocene aeolian dunes graduate downwards into beach deposits (Mastronuzzi & Sansò, 2002b). The beaches are generally less than 40-m wide and are invariably bordered by a dune belt which often covers the mid-Holocene cemented dune and by swamps which have been reclaimed during urban development (Caldara, Centenaro, Mastronuzzi, Sansò, & Sergio, 1998; Mastronuzzi, Palmentola, & Sansò, 2002). Residual dune belts, which constitute the main reservoir of sand for the natural nourishment of beaches, are under severe pressure from erosion. In addition, the beaches are fed by shore transport flooding in a south-easterly direction, rather than by the very low level of bedload transport from the ‘lame’.

The shoreline is affected by the descending flow of the Adriatic marine current (Artegiani et al., 1997) and is subject to winds, frequently blowing from the NNW and the ESE directions, which influence the wave patterns. Therefore the shoreline is subject to higher waves (about 5 m), blowing from a NE direction, as consequence of low-frequency storm waves.

3. Methods

In order to evaluate the erosion hazard along the Capitolo coast, this study collates and presents the main features influencing coastal evolution and conservation. A preliminary review of coastal hazard assessment methods was carried out (De Pippo et al., 2008; Mahendra, Mohanty, Bisoyi, Srinivasa Kumar, & Nayak, 2011; Nunes et al., 2009; Pendleton, Thieler, & Jeffress, 2005). Given the intended audience of this study, hazard levels were assessed using the methodology of the Regional Coastal Plan (Regione Puglia, 2007), with improvements made in order to reflect the characteristics of the study area. Hazard is estimated by means of a matrix representing an array of four environmental factors: exposure to high-intensity storm waves, dune preservation, shoreline evolution trends in the previous period and current shoreline evolution trends. The influence of each factor is given a value for stretches of shore about 25 meters long and the weight is arranged in the matrix according to the analysis carried out.

Coastal trends were assessed using the Digital Shoreline Analysis System 3.2 (DSAS; http://woodshole.er.usgs.gov/project-pages/dsas/). DSAS allows users to calculate shoreline changes and rate-of-change statistics from multiple historic shoreline positions (Thieler, Himmelstoss, Zichichi, & Ergul, 2009). Data processing is computed on intersection points between multitemporal shoreline positions and transects. These transects are arranged perpendicular to a user-defined baseline which is established as adjacent to the series of shoreline positions. Spacing between transects is a user-defined feature, set according to coastline morphology and length. The distances from the baseline to each intersection point along the transect are used to compute the statistics.

Multiple shoreline positions were drawn based on the ortophotos available online at the National Geoportal Web Map Service (Ministero dell'Ambiente, 2011) from 1994, 1997, 2000, 2005, 2006. In addition, we accessed the 2007 and 2010 shoreline files, available online at the Apulia Geoportal website (Regione Puglia, 2011).

The accuracy of a mapped shoreline is subject to a range of measurement and survey errors, such as residual georeferencing, seasonal shoreline variation, distortion of the photographs because of relief displacement, and inaccuracies in identifying the land-water interface (Milli & Surace, 2011; Moore, 2000). As a result, the mean multitemporal shoreline position uncertainty is about ±2.5 meters.

Shoreline positions for 1994–2005 were surveyed to study the shoreline evolution trends for that period, and 2006–2010 data were used to establish the current shoreline evolution trend. The DSAS tool was computed for both periods with transect spacing of 25 meters, giving a total of 95 transects for the Capitolo beach. Net Shoreline Movement and End Point Rate, two of the rate-change statistics provided by DSAS, were used to quantify retreating and advancing trends along the coastline. The Net Shoreline Movement (NSM) reports the distance between the oldest and youngest shorelines for each transect. The End Point Rate (EPR) is calculated by dividing the distance of shoreline movement by the time elapsed between the oldest and the most recent shoreline measures. Positive values of both NSM and EPR indicate an advancing shoreline while negative values indicate a retreating shoreline.

In order to assess dune preservation we used the official coastal survey project data. Data on stable and eroded dunes are available on the Apulia Geoportal website (Regione Puglia, 2011). We also carried out field surveys to improve the distinction between stable and eroded dunes.

An analysis of the marine weather conditions was carried out, based on data recorded by a buoy of the National Wave Measuring Network (Rete Ondametrica Nazionale, ISPRA IDROMARE, 2007) moored off the coast of Monopoli (17° 22′ 36.1″ E, 40° 58′ 30.0″ N; http://www.idromare.it). Comparisons between wind direction data (°N) and significant spectral wave height data (m) allow calculation of the wind direction that creates the highest waves (in this case NE, with the highest waves at about 5 meters). Since the Capitolo beach has a north-easterly aspect, the highest waves impact all along the coast under study.

In the environmental matrix the following weights were assigned:

1.	Retreating (0) or Advancing (1) coastline stretches a. Extent of Retreat (1994–2005): 1–20 based upon Regione Puglia (2007). b. Extent of Retreat (2006–2010): 1–50 based upon Regione Puglia (2007).
2.	Dune Stability: stable (0) or eroded (1). For eroded regions a subsequent weight of 1–30 indicating the degree of erosion was further applied.
3.	Wave Height: 1–20 based upon Regione Puglia (2007).

For each 25-meter coastal stretch, the sum of factor weights, computed as described above, allows the definition of three hazard levels according to the categories stated in the Regional Coastal Plan (Regione Puglia, 2007):

C1 – high hazard level (sum of factor weights equal to or greater than 75);

C2 – medium hazard level (sum of factor weights in the range 40–75);

C3 – low hazard level (sum of factor weights less than 40).

A 150-meter buffer zone is shown on the main map at each ‘lame mouth’ indicating a ban on development; because of this restriction, no hazard levels were stated for these areas.

4. Results and conclusions

The weather data analysis shows that the most frequent wave directions are from the NNE and ESE, which corresponds to the ‘Maestrale’ and ‘Levante’ winds. However, the greatest impact on the coast is from the highest waves generated by the ‘Grecale’ wind blowing from the North East.

Regarding dune preservation, the field survey shows that only 25% of the total dune systems appear to be stable.

The DSAS analysis showed that from 1994 to 2005 most stretches of beach were advancing, with an NSM average value of 1.76 meters and a speed of 0.16 meters per year. Further, from 2006 to 2010 there was significant retreat along ∼70% of the total coastline. The most severely eroded section is between transects 497 and 522, which corresponds to the down-drift side of the development above the Capitolo beach. This development might therefore be held responsible for the lack of longshore sediment supply and for the negative sediment balance that has resulted in significant beach retreat, with an average NSM of ∼5 m. The overall average retreat along the Capitolo sandy coast is about 0.9 m, with a mean rate of retreat of almost 10 cm per year and a peak of 47 cm per year along the most critical section. Taking into account restricted areas at ‘lame mouths’, analysis of the environmental matrix data led to the conclusion that almost 10% of the beach length is at the ‘low hazard’ level, about 67% at the ‘medium hazard’ level and nearly 23% at the ‘high hazard’ level.

Software

Data analysis and map layout were performed using ESRI ArcGIS 9.3. Transects were generated by GIS extension DSAS (version 3.2). Graphs were created using Microsoft Excel.

Acknowledgements

The authors wish to thank Giovanni Vitofrancesco, Sergio De Feudis and Raffaella Bologna, of the ‘Servizio Demanio e Patrimonio’ of the government of Apulia, for their kind collaboration and interesting contributions to discussions.