Please use this identifier to cite or link to this item: http://hdl.handle.net/2307/4560
Title: Physical and numerical modelling of landslide-generated tsunamis at a conical island : generation, propagation and early warning
Authors: Romano, Alessandro
metadata.dc.contributor.advisor: Franco, Leopoldo
Keywords: tsunamis
conical island
edge waves
tews
Issue Date: 5-Jul-2013
Publisher: Università degli studi Roma Tre
Abstract: Tsunamis are natural phenomena that can produce devastating e ects for life and human activities. They can be generated mainly by earthquakes or landslides. As far as tsunamis generated by landslides that occur directly at the coast are concerned, it has to be noted that the gererated waves can propagate both seaward than along the shoreline itself. Therefore, several scienti c and engineering issues arise. Indeed, the waves can be trapped by the bathymetry, inundating the coast itself for long distances. Furthermore, the waves that propagate seaward can reach the facing coasts (if any) with devastating e ects. In this work landslide-generated tsunamis that occur at the anks of a coni- cal island have been studied. New laboratory experiments have been carried out in a large wave tank to gain insight on tsunami generation and propa- gation mechanisms. The data collected during these new experiments are intended to be a benchmark dataset for validating analytical and numerical models. This has been achieved using a new acquisition technique, that allowed to obtain a large number of repetitions for each experiment along with a high spatial resolution of the measurements. The results of the new experiments are herein presented and discussed in depth. A detailed analy- sis of wave generation is provided. The in uence of the landslide thickness is evaluated, since two di erent landslide models have been used during the experiments. The repeatibility of the experiments is statistically quanti ed. Moreover, the features of the generated waves, both near the generation area and around the island, are described. Given the large number of time series collected around the island, a study of the spatial structure of the wave eld has been carried out, and it is pre- sented in this work. The k-f analysis has been applied in order to identify the dispersion relation followed by the waves that propagate along the shoreline (i.e., run-up). Furthermore, this technique allowed to study in detail the physics of wave propagation around the island. The relevance of each wave mode, which occurs during the propagation, is then adequately discussed. Furthermore, aiming at improving the tsunamis early warning systems in the far eld (TEWS), the application of a numerical model, which is based on the mild-slope equation (MSE) solved in the frequency domain, is pre- sented. The method proposed herein takes advantage of an inversion tech- nique which can be used in real time to reconstruct the tsunami waveform in the far- eld. The method is e ective in reconstructing the free surface el- evation time series during the tsunami event. Consequentely, this technique seems to be suitable to improve the tsunamis early warning systems. Finally, aiming at providing experimental tools to study the propagation phenomena of tsunamis, it is shown in this work a simple approach to im- prove the generation technique of solitary waves in experimental tests. Soli- tary waves are often used in laboratory experiments to study the propaga- tion and the interaction with the coasts of tsunamis. A correction technique, that aims at minimizing the discrepancies between the experimental pro le and the theoretical one, is herein presented. The technique is shown in the Appendix of this work.
URI: http://hdl.handle.net/2307/4560
Access Rights: info:eu-repo/semantics/openAccess
Appears in Collections:T - Tesi di dottorato
Dipartimento di Ingegneria

Files in This Item:
File Description SizeFormat
PhD_Thesis_Alessandro_Romano.pdf25.47 MBAdobe PDFView/Open
SFX Query Show full item record Recommend this item

Page view(s)

8
Last Week
0
Last month
0
checked on Sep 20, 2020

Google ScholarTM

Check


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.