Evaluation of the tsunamigenic potential of the largest active faults in the Alboran Basin (westernmost Mediterranean Sea)

The Alboran Basin located at the western Mediterranean hosts the boundary between the European and African plates. Previous studies support that seismicity is constrained to crustal depths. Thus, recent data acquired in the Alboran Basin during the last 15 years are key to constrain the crustal thickness as well as to characterize the sedimentary and tectonic structure of the area. The outcome of these successive surveys allows us to characterize with detail the seismic parameters of the main active faults systems (i.e. Carboneras, Al-Idrissi, Yusuf and Alboran Ridge), and to reveal their seismogenic and tsunamigenic potential.

To estimate the tsunamigenic potential for each fault, we follow a two-step workflow: First, we analyse the faults in the area and generate 3D models of the fault planes. Our dense grid of multichannel seismic reflection data, together with selected pre-stack depth-migrated sections and a wide-angle seismic refraction profile showing the velocity structure across the basin, allows us to analyse for the first time in the area, the realistic geometry of the selected faults. Second, we calculated the possible tsunami scenarios for each fault system, using our 3D fault models and their simplified rectangular plane version, all together with their associated seismic parameter estimations. Tsunami simulations were perform by using the Tsunami-HySEA non-linear shallow water GPU-optimised code developed by the EDANYA Group of the University of Malaga, Spain (de la Asunción et al., 2013, Comput. Fluids).

Comparing the tsunami obtained with the 3D and the rectangular models, we systematically observe distinct wave propagation patterns with enhanced wave heights. Thus, our results demonstrate the importance of fault characterization to improve the definition of the tsunamigenic potential of the fault systems of the Alboran Basin, providing an example for applications in other areas. In a further analysis, we are testing the importance of the slip distribution along the fault plane (i.e., uniform slip distribution vs. slip patches stochastically distributed).