Seite - 1106 - in Book of Full Papers - Symposium Hydro Engineering

(acceleration, velocity, or displacement) of a simple oscillator to the ground motion. Table 2 Values of PGA obtained from DSHA (in fractions of g) Fault name Distance (km) M Dam site horizontal vertical 50% 84% 50% 84% Harirud 7.8 7.6 0.362 0.580 0.240 0.416 Konar- Nari 1.2 7.3 0.40 0.64 0.312 0.533 The oscillator has the same period of vibration as the fundamental period of the structure. The maximum response is plotted versus the undamped natural period or the natural frequency. Site-specific response spectra are derived from ground motions arising from distinct, well-identified seismic sources in the region considered. For Gambiri diversion dam site, different methods were chosen to calculate the specific response spectra, namely: (1) probabilistic method using the line source model, (2) deterministic method using active faults in the site area, and (3) statistical method using existing strong motion records. In the following these three methods are briefly described. 4.1. RESPONSE SPECTRA FROM LINE SOURCE MODEL Some of the attenuation laws used in the PSHA are also frequency dependent. These laws were used to establish so-called Uniform Hazard Spectra or Equal Probability Spectra. On such a spectrum curve each point has an equal probability of exceeding a ground motion parameter (acceleration, velocity, displacement). By means of the logic tree procedure weighted averages of the spectra can be derived. 4.2. RESPONSE SPECTRA FROM DETERMINISTIC MODEL This model is used for the estimation of the response spectrum for the MCL. The ground motions at a site are estimated deterministically for a selected earthquake scenario. After having determined the earthquake magnitude of a specific seismic source and the closest distance to the site, the site ground motions are estimated using ground motion attenuation laws. The response spectrum is then calculated within a certain range of periods. 50th and 84th percentile values can then also be computed for different damping values. 1106