Investigating strong ground-motion variability using analysis of variance and two-way-fit plots

A statistical method to quantitatively assess the relative importance of unmodelled site and source effects on the observed variability (σ) in ground motions is presented. The method consists of analysis of variance (ANOVA) using the computed residuals with respect to an empirical ground-motion model for strong-motion records of various earthquakes recorded at a common set of stations. ANOVA divides the overall variance (σ ²) into the components due to site and source effects (respectively σ _S ² and σ _E ² ) not modelled by the ground-motion model plus the residual variance not explained by these effects (σ _R ² ). To test this procedure, four sets of observed strong-motion records: two from Italy (Umbria-Marche and Molise), one from the French Antilles and one from Turkey, are used. It is found that for the data from Italy, the vast majority of the observed variance is attributable to unmodelled site effects. In contrast, the variation in ground motions in the French Antilles and Turkey data is largely attributable, especially at short periods, to source effects not modelled by the ground-motion estimation equations used.