A new methodology for quantification of the systematic path effects on ground-motion variability

This article describes a new approach, the path diagram, for the presentation of the path component of the error in ground-motion estimates for future earthquakes. We use 19,887 records for 150 crustal earthquakes with moment magnitudes greater than 4.0 obtained from the Taiwan Strong Motion Instrumentation Program network to build the Taiwan ground-motion prediction equations for peak ground acceleration (PGA) and spectral accelerations (SAs). The record-to-record residuals are divided into small brackets in a path diagram for six source-to-site distance bins and eight source-to-site azimuth bins. The mean residuals are estimated for each path bin, from which we can get 48 path-to-path residuals for a site and compute a repeatable path term for all residuals of all stations. The remaining unexplained residuals are then estimated through the decomposition of the variance components. Comparing the results with those obtained with the same data, but using the closeness index approach, shows that we get a lower remaining variance and the higher repeatable path term with the new method. For PGA and SAs at periods of 0.3, 1.0, and 3.0 s, the remaining unexplained intraevent standard deviations obtained in the path diagram approach are 29%-40% smaller than the record-to-record standard deviation.