Stafford PJ. 2022. A model for the distribution of response spectral ordinates from New Zealand crustal earthquakes based upon adjustments to the Chiou and Youngs (2014) response spectral model. Lower Hutt (NZ): GNS Science. 97 p. (GNS Science report; 2022/15). doi:10.21420/5098-0S19.
Abstract
A model is presented for predicting the distribution of logarithmic response spectral ordinates from crustal earthquakes in New Zealand. The model includes predictions of the mean logarithmic spectral amplitudes, the logarithmic standard deviation and epistemic uncertainty in both of these components. Three logic-tree branches are proposed for both the mean and standard deviation, leading to nine logic-tree branches overall. The weights assigned to each branch are provided as part of the model specification. The model begins with the Chiou and Youngs (2014) ground-motion model and derives adjustments to that model based upon apparent differences in Fourier spectral parameters. Epistemic uncertainties are then accounted for through consideration of host-to-target differences in Fourier spectral parameters, functional form uncertainties and the extent of empirical constraint. The final model allows for predictions for all types of rupture scenarios relevant for New Zealand crustal earthquakes. However, strictly, the ranges of applicability in magnitude-distance space are approximately M4.5–8.4 and RRUP from 0 to 300 km. The applicability of shear-wave-velocity scaling is assumed to be the same as that of the Chiou and Youngs (2014) model (although the New Zealand data does not cover this range), and depth-to-top-of-rupture scaling is assumed valid from 0 to 20 km. Predictions are valid for periods between 0.01 and 10 seconds, inclusive. The model is parameterised in such a way that extrapolation beyond these ranges does not lead to significant issues. However, the small-magnitude New Zealand data possesses some unusual characteristics, so comparisons with recordings from magnitudes less than M4.5 are questionable. (The author)
