ADdGROUND

Currently the seismic hazard in a specific location e.g. Finland can be defined using a combination of locally available earthquake catalogues listing the historic earthquakes and empirical attenuation relationship calibrated based on strong motion records from areas geologically similar to Fennoscandia. The earthquake catalogues are used to estimate the probability of certain magnitude earthquakes happening in a given time period; while attenuations relationships are used to estimate how accelerations travel from the hypocenter to an observation point. Because higher magnitude earthquakes occurred before the instrumental period, no near-field strong motion recordings exists from Fennoscandian sources, so estimates of attenuation cannot be based on local data (except for very low magnitude events). Findings from the “Seismic Safety of NPP’s -Targets for Research and Education (SESA)” project highlighted that most of the hazard to a site in Finland is generated from near-field (<50km), mid-magnitude

(ML3-5) seismic events (Fülöp et al. 2012). However, predictions using attenuation relationships in the proximity of the epicenter areas are very uncertain. This is because in these regions, the distance to the epicenter or the travel path plays little role in defining the level of acceleration, more important are the properties of the earthquake source.

Within this project we propose to develop earthquake source modeling using state of the art and self-developed software in order to better understand the ground vibrations in the areas close to the earthquake epicenters. The results will enable improved consideration of earthquake hazard, in particular the effect of near-field earthquakes, thereby making nuclear facilities and hazardous industrial buildings (and other structures) safer.