Improvement of a coastal vulnerability index and its application along the Calabria Coastline, Italy.

The present paper further develops a coastal vulnerability index formulation (CVI) previously proposed by the authors by integrating a new variable and redefining three variables to improve the suitability of the index for low-lying coasts. Eleven variables are divided into three typological groups: geological, hydro-physical process and vegetation. The geological variables are: geomorphology, shoreline erosion/accretion rates, coastal slope, emerged beach width, and dune. The hydro-physical process group includes: river discharge, sea-level change, mean significant wave height and mean tide range. The vegetation variables are: vegetation behind the back-beach and coverage of Posidonia oceanica. The index was applied to a stretch of the Ionian coast in the province of Crotone in the Calabria region (Southern Italy), and a vulnerability map was produced. A geography information system (GIS) platform was used to better process the data. For the case study area, the most influential variables are shoreline erosion/accretion rates, coastal slope, emerged beach width, dune, vegetation behind the back-beach, and coverage of Posidonia oceanica. The most vulnerable transects are those near urban areas characterized by the absence of dunes and vegetation. Statistical and sensitivity analyses were performed, and the proposed CVI was compared with the previous formulation proposed by the authors and with two other CVI methods present in the literature.

Multivariate sea storm hindcasting and design: the isotropic buoy-ungauged generator procedure.

The present work provides indications for assessment of wave climate and design of structures at sea at ungauged sites, both critical issues in Ocean sciences. The paper is of methodological nature and of global worldwide applicability. It shows how suitable wave hindcasting relations can be exploited in order to provide sea storm scenarios at an ungauged (Target) location useful for design purposes: in particular, only geographical information and the knowledge of another gauged (Source) buoy are used. Several are the novelties introduced. (i) New hindcasting relations are derived. (ii) A full statistical model is set up for the Target area, whereas traditional hindcasting simply transfers time series from a gauged to an ungauged site: this gives the possibility to appropriately deal with design and hazard assessment at the Target location. (iii) The multivariate behavior of non-independent random variables is properly modelled by using the Theory of Copulas. As an illustration, a number of case studies is investigated, involving four pairs of buoys which, given their positions and exposures, are representative of a wide variety of sea states and conditions, as well as of different wave generation mechanisms.