ductility structures, and an amplification factor of 2.5 is used, the seismic base shear coefficient for design would be 0.15 ~ 0.18. As shown in Figure 4.14, the spectra indicate that low ductility structures (=2) were subjected to lateral force coefficients in the region of 0.6-0.7, about 4-5 times as much as the code coefficient. Even for long period structures, the seismic code coefficient from the calculated spectra is about 0.15 or more, much higher than the code would have indicated. Therefore, even if these structures were designed to resist seismic forces according to the code, they would have suffered unexpectedly high levels of damage.0.8 0.7 Acceleration Spectra (g) 0.6 0.5 0.4 0.3 0.2 0.1 0 0 0.5 1 1.5 Period (sec) 2 2.5 3=2 =3 =4 UBC 97 with SD UBC 97 with SBFigure 4.14 Comparison with seismic coefficient of design spectra by UBC 97Two attenuation relationships by Ambraseys et al (2005) and Campbell et al (2003) are employed to assess the severity of motion in this earthquake, and to construct iso-acceleration plots. The latter two relationships are selected because they pertain to strike-slip and thrust mechanisms, large magnitude, and a large and uniformly processed data base. In this report, two soil types, i.e. soft and stiff soil, are used as shown in Table 4.4. For example, firm and very firm soils in the attenuation relationship by Campbell et al (2003) are jointly categorized as soft soil. Figure 4.15 and Figure 4.16 illustrate the attenuation of ground acceleration with standard deviation for thrust and strike slip faults measured on soft soil for horizontal and vertical ground motion. The peak ground acceleration values from YOGI and BJI stations are also shown along with error bars corresponding to lower and upper limits established from earlier sections of this report. Distances are measured from the presumed epicenter by NIED due to uncertainty in fault rupture location and length. The selected attenuation relationships support the PGA estimated by back-analysis and reconstructed velocity records. The attenuation for strike-slip mechanisms tends to have a good agreement with PGAs at YOGI and BJI, especially, the attenuation by Campbell et al (2003).Table 4.4 Soil class for each attenuation relationshipSoil Class Soft Soil Stiff Soil Ambraseys et al (2005) Soil type Shear Velocity (m/s) Soft soil Stiff soil 180 ~ 360 360 ~ 750 Campbell et al (2003) Soil type Shear Velocity (m/s) Firm soil 210 ~ 390 Very firm soil 290 ~ 490 Soft rock 310 ~ 530 Firm rock 490 ~ 1170Mid-America Earthquake Center21
