GNGTS 2016 - Atti del 35° Convegno Nazionale

330 GNGTS 2016 S essione 2.1 the tsunami produced some evident signals needs to be investigated. One could argue some resonance effects took place in the harbor, specifically related to the basin dimensions that enhanced considerably the oscillations coming from the open sea. Indeed, the tsunami produced by the earthquake, according to the numerical simulations, is very small (in the order of a few cm in the open sea) and reached not only the Crotone port but many other similar ports of Greece and South Italy. So the question is: why a similar anomaly was noticed only in Crotone? Other ports have similar dimensions, but the tide gauges did not show any specific oscillations. One could try to explain the anomaly by invoking a source that affects only Crotone, and that is therefore a local tsunami source, such as, for example, a small submarine landslide. These issues are investigated here by means of in-house developed codes: the earthquake- generated tsunami is simulated by a finite difference resolution scheme (implemented in UBO- TSUFD code), solving the shallow-water equations over a regular grid representing the sea depth, and accounting also for the wave impact on the coast. The nested grids technique allows one to zoom over areas of interests, describing local features with a better resolution: in this case, the larger grid covers the whole Ionian Sea, from the source located in Eastern Greece to Calabria, while the port of Crotone is studied with higher resolution in order to account for resonance effects inside the harbor. As concerns the local source, a small landslide with volume of about 8 million m 3 has been hypothesized along the submarine steep margins off Crotone, where a lot of features can be observed related to previous mass collapses (scars, submarine canyons). The assumption is that the seismic shaking produced by the Lefkada earthquake was able to mobilize a small underwater mass in an area of high landslide susceptibility. Again, a self-developed and maintained code is here applied, named UBO-BLOCK1, simulating the mass motion time history and providing then the time-dependent impulse generating the tsunami. The tsunami propagation is computed bymeans of the alreadymentioned UBO-TSUFD code, and the simulated sea level signal compared with the observed one. The simulations show that the slide is able to generate directly the observed signal, meaning that at least a local contribution to the anomaly can be invoked, possibly enhancing or even provoking the resonance in the harbor. The November 17 anomaly, apart from its interest from the scientific point of view, represents an important instance for the tsunami warning issues: small perturbations, when meeting particular conditions in coastal basins, can be considerably enhanced, affecting the coast for long time, with the possibility of huge damages on ships and infrastructures. Such events are not necessarily associated with huge earthquakes, that in such cases can be considered an important precursor of the tsunami occurrence. Moreover, a better sea-level instrumental coverage of the harbors in South Italy could provide a more detailed description of the background signal, and help understand local features such as the typical resonance periods. In turn, this will provide hints on the possible local amplification of tsunamis and point out which tsunamis constitute a potential threat for the coastal community. Acknowledgements. This study is carried out and funded 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 Ganas A. et al. (2016) Coseismic deformation, field observations and seismic fault of the 17 November 2015M=6.5, Lefkada Island, Greece earthquake, http://dx.doi.org/10.1016/j.tecto.2016.08.012