Basin Edge Effect on Seismic Ground Response: A Parametric Study for Duzce Basin Case, Turkey


Hasal M. E. , İyisan R. , Yamanaka H.

ARABIAN JOURNAL FOR SCIENCE AND ENGINEERING, cilt.43, sa.4, ss.2069-2081, 2018 (SCI İndekslerine Giren Dergi) identifier identifier

  • Cilt numarası: 43 Konu: 4
  • Basım Tarihi: 2018
  • Doi Numarası: 10.1007/s13369-017-2971-7
  • Dergi Adı: ARABIAN JOURNAL FOR SCIENCE AND ENGINEERING
  • Sayfa Sayıları: ss.2069-2081

Özet

In this study, edge effect on the spatial variation of surface ground motion was evaluated by conducting one (1D) and two-dimensional (2D) dynamic analyses on the numerical model of Duzce basin. Duzce is a province in northwestern Turkey and was hit by a devastating earthquake in November 1999. In order to estimate the seismic behavior of Duzce basin, 1D and 2D dynamic response analyses were carried out on its numerical model by using the seismic bedrock motion data derived from the deconvolution analysis of the N-S component of 1999 Duzce earthquake accelerogram. In addition, a series of different accelerograms and slope values were utilized to investigate the possible effects of change in strong ground motion intensity level and edge geometry on the dynamic response of soil layers over inclined bedrock. The finite element model for the edge section of the flat sedimentary basin was set up by using the 2D shear wave velocity profile, geotechnical and geological site conditions in Duzce. The variation of 2D/1D spectral acceleration ratios and aggravation factors with distance from basin edge was obtained for different period values. The results indicate that the aggravation factors reached their maximum values at a specific zone near basin edge and they generally converged to 1 regardless of the bedrock inclination or period values with increasing distance from the edge section. The highest aggravation factor values were calculated as 2.0-2.5 for the period interval of 0.2-0.6 s. A relationship was proposed between the aggravation factor and basin edge geometry.