Modelling as a tool to augment ground motion data in regions of diffuse seismicity (AddGROUND):Final report

• Main Authors: Fülöp, Ludovic, Jussila, Vilho, Lund, Björn, Fälth, Billy, Voss, Peter H., Puttonen, Jari, Saari, Jouni, Heikkinen, Pekka
• Format: Book
• Language: English
• Published: Nordic Nuclear Safety Research NKS 2017
• Subjects: nuclear power plant safety; fault source modeling; near-field effects; earthquake; Fennoscandia
• Online Access: https://cris.vtt.fi/en/publications/8ca63550-c175-45fa-b936-c0a33c0b3588; http://www.nks.org/en/nks_reports/view_document.htm?id=111010214562502

De-aggregation of probabilistic hazard assessment (PSHA) results show that the dominating source of vibrations with engineering significance to NPP safety is from mid-magnitude earthquakes located at close distances to the plant. This region is called the “near-field” and is known for its particularities when compared to “far-field”. For example, significant duration of the ground motions is shorter, corresponding to S-wave and surface wave arrivals; there are distinctive high velocity peaks in the ground motions and vertical shaking components may exceed horizontal components. These particularities are known to have design consequences, but are often overlooked by engineering codes. In Fennoscandia, near-field observations of larger magnitude (M>3) earthquakes are missing, and modelling is the only way to supplement the existing empirical data underspinning the attenuation equations in the PSHA studies. During the project year 2015, we confirmed the near-source effect in small magnitude earthquake recordings in Finland and developed modeling skills and tools to generate synthetic, near-field accelerograms starting from process of the fault rupture. Within this report (2016), we describe the modelling techniques and compare the modelling outcomes for Mw=5.5 earthquakes with ground motion prediction equations GMPE’s developed for stable continental regions. Five cases were analyzed in order to explore the capabilities of ground motion simulation tools. In the five cases, the varied parameters were depth of the source, dip angle of the fault and dynamic properties of the fault. The models were developed in COMPSYN in 3DEC. By these comparisons we highlight the potentials and limitations of modelling to support empirical observations.