A Predictive Ground-Motion Model for Turkey and Its Comparison with Recent Local and Global GMPEs

Abstract

AbstractWe present a local ground motion prediction equation (GMPE ) for estimating the peak ground-motion values that was derived using the recently compiled Turkish strong motion database . The new GMPE is comparable with the recent global GMPEs in terms of model sophistication and quality of underlying database. Using this equation, we explore the inter-event , inter-station ground motion variability of the recent Turkish strong motion database as well as suitability of some local and global GMPEs for regional seismic hazard assessment analyses. The inter-event error underlines the distinguishing characteristics of few earthquakes, suggesting that the use of local GMPEs can be important especially when specific scenario studies are to be carried out. The inter-station variability allowed us to detect stations with outlier site response and to investigate the goodness of the employed site effect s model. Results also indicate suitability of some global and local GMPEs for use in regional seismic hazard assessments together with the model presented in this study. These should be combined through a logic tree scheme to reduce both aleatory and epistemic uncertainty in local hazard assessments.KeywordsGround MotionSeismic HazardPeak Ground AccelerationStrong MotionSeismic Hazard AnalysisThese keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.