Simulation of large earthquakes and its implications on earthquake insurance rates: a case study in Bursa region (Turkey)

Ünal B., Askan A., Selcuk-Kestel A. S.

NATURAL HAZARDS, vol.85, no.1, pp.215-236, 2017 (SCI-Expanded) identifier identifier

  • Publication Type: Article / Article
  • Volume: 85 Issue: 1
  • Publication Date: 2017
  • Doi Number: 10.1007/s11069-016-2578-4
  • Journal Name: NATURAL HAZARDS
  • Journal Indexes: Science Citation Index Expanded (SCI-EXPANDED), Scopus
  • Page Numbers: pp.215-236
  • Middle East Technical University Affiliated: Yes


Ground motion intensity parameters of past and potential earthquakes are required for a range of purposes including earthquake insurance practice. In regions with no or sparse earthquake recordings, most of the available methods generate only peak ground motion parameters. For cases where full ground motion time histories are required, simulations that consider fault rupture processes become necessary. In this study, a major novel use of simulated ground motions is presented in insurance premium calculations which also require ground motion intensity measures that are not always available through observations. For this purpose, potential earthquakes in Bursa are simulated using stochastic finite-fault simulation method with dynamic corner frequency model. To ensure simulations with reliable synthetic ground motions, input parameters are derived from regional data. Regional model parameters are verified by comparisons against the observations as well as ground motion prediction equations. Next, a potential large magnitude event in Bursa is simulated. Distribution of peak ground motion parameters and time histories at selected locations are obtained. From these parameters, the corresponding Modified Mercalli Intensities (MMI) are estimated. Later, these MMIs are used as the main ground motion parameter in damage probability matrices (DPM). Return period of the scenario earthquake is obtained from the previous regional seismic hazard studies. Finally, insurance rates for Bursa region are determined with implementation of two new approaches in the literature. The probability of the scenario event and the expected mean damage ratios (MDR) from the corresponding DPMs are used, and the results are compared to Turkish Catastrophe Insurance Pool (TCIP) rates. Results show that insurance premiums can be effectively computed using simulated ground motions in the absence of real data.