302

GNGTS 2017 SeSSione 2.1

on the use of fAults And bACkGround seIsmICIty In seIsmIC probAbIlIstIC tsunAmI hAzArd AnAlysIs (spthA) J. Selva1, S. Lorito2, A. Babeyko3, R. Basili2, A. Hoechner3, F.E. Maesano2, Antonio Scala2, M. Taroni2,R. Tonini2, M.M. Tiberti2, F. Romano2, P. Perfetti1, M. Volpe2

1 Istituto Nazionale di Geofisica e Vulcanologia, Bologna, Italy2 Istituto Nazionale di Geofisica e Vulcanologia, Roma, Italy3 GFZ German Research Centre for Geosciences, Potsdam, Germany

Most of the SPTHA studies and applications rely on several working assumptions: i) the mostly offshore tsunamigenic faults are sufficiently well known; ii) the subduction zone earthquakes dominate the hazard; and iii) their location and geometry is sufficiently well constrained. Hence, a probabilistic model is constructed as regards the magnitude-frequency distribution and sometimes the slip distribution of earthquakes occurring on assumed known faults. Then, tsunami scenarios are usually constructed for all earthquake locations, sizes, and slip distributions included in the probabilistic model, through deterministic numerical modelling of tsunami generation, propagation, and impact on realistic bathymetries.

Here, we adopt a different approach (Selva et al., 2016) that relaxes some of the above assumptions, considering that: i) non-subduction earthquakes may also contribute significantly to SPTHA, depending on the local tectonic context; ii) not all the offshore faults are known or sufficiently well constrained; and iii) the faulting mechanism of future earthquakes cannot be considered strictly predictable from the local tectonic setting.

GNGTS 2017 SeSSione 2.1

303

Nevertheless, this approach uses as much as possible the information from known faults which, depending on the level of knowledge and on the local tectonic complexity, among other things, can be modelled in three different ways: 1) Predominant Seismicity (PS), 2) Background Seismicity (BS), Special Background Seismicity (SBS). PS is used when it is possible to assume the fault geometry and mechanism to be sufficiently known (e.g. for the main subduction zones) and future earthquakes are predicted to occur only on those fault surfaces. The BS is used for modelling the seismicity around the PS faults and in all cases where future earthquakes can be assumed to occur anywhere in a volume, similarly to typical seismic hazard approaches. In this latter approach, the potential sources are distributed on a regular lattice where the information on faults is merged with that on past seismicity, dominant stress regime, and tectonic characterization, to determine a probability density function for the faulting mechanism. The SBS is a special category used when major structures are thought to be present but their geometry and position is not well-enough constrained.

To illustrate the methodology and its impact on the hazard estimates, we present the results of an application for estimating the tsunami impact along the coastlines of the NE Atlantic, the Mediterranean, and connected Seas (NEAM) region.

For the deterministic numerical modeling of tsunamis we use the database of pre-calculated tsunami scenarios of elementary sources (Molinari et al., 2016) constructed with numerical simulations performed with Tsunami-HySEA (Macías et al., 2017). The impact of tsunamis on

Fig. 1 - Disaggregation per cell for intensity ψ> 1 m; this map is a spatial distribution of the causative source probability for a tsunami at the site which exceeds the selected threshold. Red crosses indicate the location of the target in each panel (Selva et al., 2016).

304

GNGTS 2017 SeSSione 2.1

the coastline is estimated by applying empirical amplification factors (Glimsdal et al., 2017). All the calculations are performed through the hazard calculation platform developed at INGV in collaboration with GFZ.

The results of the application consist of: i) two sensitivity analyses (on some alternative SPTHA formulations, and on the separation between BS and PS); ii) SPTHA results along selected coastlines for an exposure time of 50 yr, along with disaggregation analyses (Fig. 1), and relative uncertainty. This methodology was initially designed during the ASTARTE project and has been later applied for the regional-scale SPTHA in the TSUMAPS-NEAM project.Acknowledgements We acknowledge financial support by the EC FP7 ASTARTE (Grant agreement 603839). The TSUMAPS-NEAM Project (http://www.tsumaps-neam.eu/) is co-financed by the European Union Civil Protection Mechanism, Agreement Number: ECHO/SUB/2015/718568/PREV26.

ReferencesGlimsdal S, Lovholt F, Harbitz C, Orefice S, Romano F, Brizuela B, Lorito S, Hoechner A, Babeyko A (2017)

Development of Local Amplification Factors in the NEAM Region for Production of Regional Tsunami Hazard Maps, in Geophysical Research Abstracts, Vol. 19, EGU2017-7657-1.

Macías, J., Castro, M.J., Ortega, S., Escalante C., González-Vida J. M. (2017) Performance Benchmarking of Tsunami-as, J., Castro, M.J., Ortega, S., Escalante C., González-Vida J. M. (2017) Performance Benchmarking of Tsunami-ález-Vida J. M. (2017) Performance Benchmarking of Tsunami-HySEA Model for NTHMP’s Inundation Mapping Activities. Pure Appl. Geophys., doi: 10.1007/s00024-017-’s Inundation Mapping Activities. Pure Appl. Geophys., doi: 10.1007/s00024-017-s Inundation Mapping Activities. Pure Appl. Geophys., doi: 10.1007/s00024-017-1583-1.

Molinari I, Tonini R, Lorito S, Piatanesi A, Romano F, Melini D, Hoechner A, Gonzàlez Vida JM, Maciás J, Castro MJ, de la Asunción M (2016). Fast evaluation of tsunami scenarios: uncertainty assessment for a Mediterranean Sea database, Nat. Hazards Earth Syst. Sci., 16, 2593-2602, doi:10.5194/nhess16-2593-2016.

Selva J., Tonini R., Molinari I., Tiberti M.M., Romano F., Grezio A., Melini D., Piatanesi A., Basili R., Lorito S. (2016). Quantification of source uncertainties in Seismic Probabilistic Tsunami Hazard Analysis (SPTHA). Geophys. J. Int., 205, 1780-1803, doi:10.1093/gji/ggw107.