GNGTS 2016 - Atti del 35° Convegno Nazionale

280 GNGTS 2016 S essione 2.1 For each of the 32 sources, the tsunami initial condition was computed by means of the Okada (1985) analytical formulae, holding for a rectangular fault embedded into a perfectly elastic, homogeneous and isotropic half-space.The tsunami propagation and impactwas simulated through the UBO-TSUFD code (Tinti and Tonini, 2013) on a set of properly nested finite-differences grids. The grid-nesting facility implemented in UBO-TSUFD allows one to investigate tsunami impact on natural and man-made coastal features with detail limited by the sole resolution of the available topography and bathymetry. In the case of Augusta and Siracusa, we used a set of five nested grids with resolutions of 3 km (grid 1, covering a large portion of the Ionian sea ranging from eastern Sicily to western Greece), 1 km (grid 2, covering only western Ionian), 200 m (grid 3, along the entire eastern Sicily coast) and 40 m (grid 4: Augusta; grid 5: Siracusa). Performing the tsunami simulations for each of the 32 faults over the entire set of nested grids would be not only severely demanding from a computational point of view, but also of limited sense from a methodological perspective. In general, the goal must be to recognise those faults in the selected ensemble whose ensuing tsunami impact on the target areas is expected to be the largest. This can be accomplished by developing a suitable selection algorithm: we rank the faults based on the maximum water elevation computed on each of the coastal nodes in grid 3 (200-m resolution), and hence on the relative frequency of their appearance in the first n positions (e.g. n =10) of the ranking for all nodes. This approach is first applied to the entire eastern Sicily coasts first, and then on the coastal nodes in the areas aroundAugusta and Siracusa. The results highlight that the faults that produce the largest effects at regional level are not necessarily those with the most severe effects in the target areas. Moreover, despite the limited spatial distance between Augusta and Siracusa, the selection procedure can lead to slightly different results for the two districts. As a result of the selection approach, we are able to single out the sources expected to produce the largest tsunami effects in the target areas, reducing the number of faults of interest from 32 to only 9, for which full-resolution inundation modelling is performed. The most synthetic and most useful way to summarize the results produced in the frame of our scenario-based approach is to combine all the information coming from the single scenarios into a unique aggregated scenario. Building an aggregate scenario for a given tsunami physical variable means selecting for each grid node the maximum value of that variable among the individual scenarios. The simulation results presented and discussed in this contribution consist of fields of maximum water elevation, of maximum water column, of maximum velocity and of maximum momentum flux obtained after aggregating the outputs relative to the 9 individual tsunamigenic faults.Among the main results relative to the maximumwater elevation aggregated field, it is worth noting that: 1) the entire Augusta-Siracusa district is interested by significant wave impact, with local variations determined by the local coastal morphology or by the presence of coastal structures, such as the three breakwaters protecting the inner Augusta Bay; 2) the areas were largest inundation extents are predicted coincide with the zones were paleotsunami evidences have been found and discussed in the literature (e.g. De Martini et al. , 2012). It is the case, for example, of the area north-east of the Augusta peninsula and of the Priolo site; 3) the largest inundation are produced by the far-field faults located in correspondence with different portions of the WH arc. Acknowledgements. The research presented in this paper was carried out in the frame of the EU Project called ASTARTE - Assessment, STrategy And Risk Reduction for Tsunamis in Europe (Grant 603839, 7th FP, ENV.2013.6.4-3). References Basili R., and other 28 co-autho De Martini P. M., Barbano M. S., Pantosti D., Smedile A., Pirrotta C., Del Carlo P., Pinzi S.; 2012: Geological evidence for paleotsunamis along eastern Sicily (Italy): an overview . Nat. Hazards Earth Syst. Sci., 12 , 2569-2580.