Search

The GEBRA '93 (Geological Evolution of the Bransfield Basin) survey was carried out in December 1993 in the Bransfield Strait (NW Antarctic Peninsula) and adjacent areas, onboard BIO Hespérides. The geophysical data acquired allow to characterize the morphostructure, sediment cover and magnetic anomalies of the SW-NE trending Central Bransfield Basin. Some transverse shifts and tresholds, together with alongbasin volcanic alignments, complicate the morphology and seismic structure of what otherwise would have been a long, continuous axial trough opened on a crust under extension. A major submarine valley, named Gebra Valley has also been identified, Sí

This article presents a reasonable present-day, sea-level highstand numerical simulation and scenario for a potential tsunami generated by a landslide with the characteristics of the BIG’95 debris flow, which occurred on the Ebro margin in the western Mediterranean Sea in prehistoric times (11,500 cal yr BP). The submarine landslide deposit covers an area of 2200 km2 of the slope and base of slope (200–1800-m water depth), involving a volume of 26 km3. A leapfrog finite difference model, COMCOT (Cornell multigrid coupled tsunami model), is used to simulate the propagation of the debris-flow-generated tsunami and its associated impact on the nearby Balearic Islands and Iberian coastlines. As a requisite of the model, reconstruction of the bathymetry before the landslide occurrence and seafloor variation during landsliding have been developed based on the conceptual and numerical model of Lastras et al. (2005). We have also taken into account all available multibeam bathymetry of the area and high-resolution seismic profiles of the debris flow deposit. The results of the numerical simulation are displayed using plots of snapshots at consecutive times, marigrams of synthetic stations, maximum amplitude plots, and spectral analyses. The obtained outputs show that the nearest shoreline, the Iberian coast, would not be the first one hit by the tsunami. The eastward, outgoing wave would arrive at Eivissa Island 18 min after the triggering of the slide and at Mallorca Island 9 min later, whereas the westward-spreading wave would hit the Iberian Peninsula 54 min after the slide was triggered. This noticeable delay in the arrival times at the peninsula is produced by the asymmetric bathymetry of the Catalano-Balearic Sea and the shoaling effect due to the presence of the wide Ebro continental shelf, which in addition significantly amplifies the tsunami wave (19 m). The wave amplitudes attain 8 m in Eivissa, and waves up to 3 m high would arrive to Palma Bay. Resonance effects produced in the narrow Santa Ponc¸a Bay in Mallorca Island could produce waves up to 9 m high. A similar event occurring today would have catastrophic consequences, especially in summer when human use of these tourist coasts increases significantly. This research was supported by the TRANSFER project, European Commission (EC) contract 037058-TRANSFER funded by the EC’s Sixth Framework Programme, GRACCIE-CONSOLIDER project (ref. CSD2007-00067) of the Spanish Plan Nacional I D I, and a Generalitat de Catalunya “Grups de Recerca Consolidats” grant (2009 SGR 1305). O. Iglesias and R. Durán are supported by an FPU fellowship of the Spanish Ministerio de Educación and a Juan de la Cierva fellowship of the Spanish Ministerio de Ciencia e Innovación, respectively, and E. Tahchi is supported by a Marie Curie Intra-European Fellowship. We would also like to thank all scientists and crew who participated in seagoing activities to obtain geophysical data. We thank G. Papadopoulos, an anonymous reviewer, and the editor for their thoughtful revisions. RTD projects that funded the cruises are also acknowledged. The publication reflects only the views of the authors. The EC is not liable for any use that may be made of this article.