Physically-Based Ground Motion Prediction and Validation: A Case Study of Mid-Sized Marmara Sea Earthquakes

Abstract

Computation of realistic time histories for different locations around Marmara region can be helpful for engineering design, retrofitting the existing structures, hazard and risk management studies and developing new seismic codes and standards. This paper had two main purposes. The first one was to simulate five moderate earthquakes (Mw ≈ 5.0) recorded in the Marmara region. We synthesized ground motion for the full wave train on three components, and applied a ‘physics based’ solution of earthquake rupture. For each earthquake, we synthesized seismograms using 500 different rupture scenarios that were generated by Monte Carlo selection of parameters within the range. The second purpose was to validate synthetic seismogram with real seismogram. To develop credibility of a synthetic seismogram in engineering point of view, we followed the methodology of Anderson (13th World Conference on Earthquake Engineering, 2003 [1]). Because this methodology produces source and site specific synthetic ground motion time histories and goodness-of-fit scores of obtained synthetics was between ‘good’ to ‘excellent’ range based on Anderson’s score. We concluded that it can be used to produce ground motion that has not previously been recorded during catastrophic earthquakes.