Systematic comparison on the inundation response of AnuGA and COMCOT Tsunami modelling codes applied to the Boca do Rio and Alvor Bay area

• Author(s): Fernandes, Miguel André Gonçalves
• Date: 2009
• Persistent ID: http://hdl.handle.net/10400.1/473
• Origin: Sapientia - Universidade do Algarve
• Subject(s): Teses; Oceanografia; Risco sísmico e de tsunami; Modelos númericos de tsunami
• Document Type: Master thesis
• Language: Portuguese
• Advisor(s): Luis, Joaquim
• Contributor(s): Sapientia

Description

Dissertação mest., Oceanografia, Universidade do Algarve, 2009

This work is integrated in the Portuguese e orts to study tsunami hazard and risk on the Algarve coast and to develop emergency plans by the Portuguese Civil Protection. As part of this broader purpose, the objective of this study is to compare the inundation response of tsunami numerical models. O the models available to this study, we have selected for comparison: model COMCOT based on a nite di erence scheme to discretize non-linear shallow water wave equations (NLSWE) and model AnuGA based on the nite volume method for the evaluation of NLSWE. We started by validating the numerical codes with analytical and laboratory benchmarks proposed at the 2004 Catalina Long Wave Congress. A rst test case was performed at Boca do Rio valley where historical and sedimentological data from the 1755 Lisbon tsunami was available. Its purpose was to properly con gure the numerical models to the Algarve coast by comparing the modelled results with the historical data. We have found a good agreement with the historical data, even though the run-in was underestimated. A systematic comparison of the inundation response of the numerical models was performed at Alvor Bay. The response was measured by the run-up and run-in parameters. We have constructed 81 fault models to create the initial conditions for the tsunami codes based on the variation of fault parameters: dip and depth. A good agreement was found in AnuGA and COMCOT propagation modelling. The di erences found in the run-up and run-in parameters seem to be related to the inundation methods. COMCOT has always produced higher inundation parameters than AnuGA. Nevertheless, the results obtained with AnuGA have been more consistent. The maximum inundation on the cases teste have been obtained with proximate source models while COMCOT has revealed a large span of source models able to produce maximum inundation.