Search

New models are developed for the prediction of vertical-to-horizontal (V/H) spectral ratios of peak ground acceleration (PGA), peak ground velocity (PGV), and 5%-damped elastic response spectra for periods ranging from 0.01 to 10 s. The models use moment magnitude, source-to-site distance, style-of-faulting (SoF), and site classification terms as independent variables. The study is based on a well-compiled Iranian ground-motion databank, consisting of 1350 (*3) three-component accelerograms generated by 370 earthquakes. Only records with known measured time-averaged shear-wave velocity in the uppermost 30 m (VS30) are used for regression analyses. The presented models are valid for moment magnitude ranging from 4.5 to 7.4, distance up to 200 km, and for earthquakes with reverse and strike–slip fault mechanisms. Four alternative mathematical forms are evaluated in terms of magnitude and distance in order to find the optimal functional form for reliable prediction of V/H ratios. Out of four proposed forms, two were chosen for further investigation. The point source (hypocentral and epicentral) as well as extended source (rupture and Joyner-Boore) distance metrics are employed to derive four models. The overall validity of the proposed models is examined by detailed residual analyses as well as comparison with some of the local and regional predictive V/H models. Our proposed models showed significant reduction in the total standard deviation across all periods. The proposed V/H models facilitate assessment of vertical components of response spectrum where horizontal component of response spectrum is known, using V/H spectral ratios as scaling factors.

We present an objective technique for assessing the fundamental resonance frequency ( $${\text{f}}_{0}$$ ) of a given site through horizontal-to-vertical (H/V) spectral ratio. H/V spectral ratios have been determined from 5%-damped spectral acceleration derived from strong-motion time histories recorded at accelerograph stations. The technique has been applied to an updated comprehensive strong ground-motion database from Iran. In this study, the resultant $${\text{f}}_{0}$$ values have been also evaluated through visual inspection of average H/V response spectral ratios for each station as well as detailed inspection of vertical and horizontal spectral components of all events recorded at each station. The technique has been applied to 389 stations of the Iran Strong Motion Network (ISMN), from which $${\text{f}}_{0}$$ values have been determined for 266 stations. The determined $${\text{f}}_{0}$$ values have been used to estimate time-averaged shear-wave velocity in the upper 30 m ( $${\text{V}}_{{{\text{S}}30}}$$ ), taking into account the correlation between H/V parameters and measured $${\text{V}}_{{{\text{S}}30}}$$ values at recording stations. Results and corresponding model uncertainties have been compared with other related studies conducted for Japan, Central and Eastern North America as well as a global model. In comparison with other datasets, the Iranian $${\text{V}}_{{{\text{S}}30}}$$ model shows the highest $${\text{V}}_{{{\text{S}}30}}$$ values among all models. We use the region-specific equation developed in this study to estimate $${\text{V}}_{{{\text{S}}30}}$$ values for 63 stations of ISMN where no other site characterization was available.