Evaluation of Wells and Coppersmith (1994) earthquake and fault relationships in the New Zealand context

Abstract

We assess the suitability of the earthquake and fault relationships of Wells and Coppersmith (1994) for use in New Zealand seismic hazard studies. We find that the regressions of Wells and Coppersmith (hereafter referred to as "W&C") provide underestimates of the moment magnitudes (Mw) and coseismic surface displacements of large New Zealand earthquakes, and attribute much of the discrepancy to the dataset used by W&C for their analysis. Regressions developed after addition of newly published data to the W&C dataset, and addition of data originally excluded from the W&C dataset provide closer estimates of Mw and displacement to those of the New Zealand earthquakes. The remaining discrepancies appear to be due to the New Zealand earthquake dataset comprising largely paleoearthquake and pre-instrumental data, which produce larger estimates of Mw and displacement than more recent data. Lastly, we find that the coseismic displacement per unit rupture length (proportional to stress drop) is not constant for all earthquakes, but is a decreasing function of both slip rate and the total amount of slip registered across the fault.