New Site Coefficients and Site Classification System Used in Recent Building Seismic Code Provisions

Abstract

Recent code provisions for buildings and other structures (1994 and 1997 NEHRP Provisions, 1997 UBC) have adopted new site amplification factors and a new procedure for site classification. Two amplitude-dependent site amplification factors are specified: Fa for short periods and Fv for longer periods. Previous codes included only a long period factor S and did not provide for a short period amplification factor. The new site classification system is based on definitions of five site classes in terms of a representative average shear wave velocity to a depth of 30 m (V¯s). This definition permits sites to be classified unambiguously. When the shear wave velocity is not available, other soil properties such as standard penetration resistance or undrained shear strength can be used. The new site classes denoted by letters A - E, replace site classes in previous codes denoted by S1 - S4. Site classes A and B correspond to hard rock and rock, Site Class C corresponds to soft rock and very stiff / very dense soil, and Site Classes D and E correspond to stiff soil and soft soil. A sixth site class, F, is defined for soils requiring site-specific evaluations. Both Fa and Fv are functions of the site class, and also of the level of seismic hazard on rock, defined by parameters such as Aa and Av ( 1994 NEHRP Provisions), Ss and Sl ( 1997 NEHRP Provisions) or Z ( 1997 UBC). The values of Fa and Fv decrease as the seismic hazard on rock increases due to soil nonlinearity. The greatest impact of the new factors Fa and Fv as compared with the old S factors occurs in areas of low-to-medium seismic hazard. This paper summarizes the new site provisions, explains the basis for them, and discusses ongoing studies of site amplification in recent earthquakes that may influence future code developments.

Hazard-consistent response spectra in the Region of Murcia (Southeast Spain): comparison to earthquake-resistant provisions

“…The 2003-NEHRP Provisions use a spectral shape anchored at two spectral accelerations (for short and long-periods that are extracted from hazard maps) and include building importance coefficients and non-linear, period-dependent soil amplification factors (Borcherdt 1994;Dobry et al 2000). Specifically, the 2003-NEHRP Provisions consider SA 247 5(0.2s) and SA2475(1.0s), the expected spectral accelerations of 0.2 and 1.0 with a 2% probability of exceedance in 50 years, equivalent to a return period of 2475 years.…”

“…Due to the large-scale approach used, it was unviable to perform site-specific analyses for assessing ground amplification factors, and these were obtained by averaging amplification factors proposed in the literature. Specifically, amplification factors for PGA (Table 3) account the non-linearity and period-dependency of soil response, amplification factors for spectral accelerations (Table 3) were adapted from the 2003 NEHRP Provisions (Dobry et al 2000). As the averaged PGA amplification factor calculated above does not correspond to the PGA amplification factor contained in the 2003 NEHRP Provisions, it was necessary to recalculate amplification factors for the short-period range (from 0.0 through 0.1 s).…”

Site Response in NEHRP Provisions and NGA Models

“…where superscript 'N' indicates re-normalization, F a and F v are the original, published NEHRP factors, V ref = 1050 m/s per Borcherdt (1994b) and Dobry et al (2000), and m a and m v are taken from Borcherdt (1994b) and Dobry et al (2000) (shown in Figure 4). No adjustments are made at PGA r = 0.5g due to a lack of published m a and m v values in Figure 4.…”

“…Similar considerations apply for nonlinear ground response analysis. Borcherdt (1994b) and Dobry et al (2000) describe the process by which equivalent linear and nonlinear ground response simulations were used to supplement the linear site factors in Figure 3. Suites of profiles were analyzed by Seed et al (1994) and for categories C-E using velocity profiles from sites in California and Mexico City.…”

“…The empirical amplification values shown in Figure 3 were found to be in good agreement with those derived independently by Seed et al (1994), those computed parametrically by at input ground motion levels near 0.1g, and response spectral ratios computed by Joyner et al (1994). Hence, the modeling results were used to extrapolate the inferred amplification factors to higher input peak acceleration levels of 0.2, 0.3, and 0.4 g. Borcherdt (1994b) and Dobry et al (2000) describe how the computed site factors were expressed in a linear form in log-log space as shown in Figure 4 and given by the following expressions: Page 8 Figure 5 also shows the NEHRP site factors plotted at the V s30 values for which category-based site factors were originally developed by Borcherdt (1994b), as explained previously. The NEHPR factors have some discrepancies from the regression lines, especially for F a in Category E and F v in Categories C-D. As mentioned previously, those discrepancies arose from committee decisions.…”