Empirical Model for the Prediction of Ground Motion Duration on Soft Soils

Abstract

In the civil engineering field, several attenuation relationships are proposed to estimate the ground motion duration (GMD) of earthquakes for hard and medium soil types, considering various seismic quantities. However, there is a lack of prediction equations that are directly used for soft soil sites. It is a known fact that soft soil has very adverse impacts on the seismic behavior of structures due to the amplification of displacement demands, and hence, GMD is also affected by this situation. Considering the importance of the topic, special attention is paid to the prediction of GMD of earthquakes at soft soil sites. In the study, two important topics are emphasized. First, a technique explicitly evolved to quantify the significant earthquake GMD based on the estimated time interval between 5 and 95% of the Arias intensity is presented. Second, a prediction equation for significant GMD is proposed for soft soils by using several earthquake database collections. The collected database is statistically evaluated for different earthquake parameters such as moment magnitude (Mw), closest distance to the rupture (Rrup) and average shear wave velocity for the top 30 m of soil (Vs,30), and a prediction equation is suggested using several regression coefficients based on these parameters. Consequently, the proposed prediction equation is compared with the existing models used in the literature, and the efficiency of the model is evaluated.