Seebeck, H.; Bassett, D.; Thrasher, G.P.; Kellett, R.L.; Pradel, G.J.; Viskovic, G.P.D.; Griffin, A.G.; Hill, M.P.; Tozer, B. 2025 Toward a New Zealand community velocity model. Lower Hutt, NZ.: Earth Sciences New Zealand. GNS Science report 2025/30. 64 p.; doi: 10.21420/20DE-E013
Abstract
Held by the Ministry of Business, Innovation & Employment (MBIE), New Zealand has a wealth of high-quality publicly available geological and geophysical data collected over the last 60 years. While acquired specifically for resource exploration, these existing data are an under-utilised resource that can be re purposed to support a range of new geological, geotechnical and geophysical research underpinning geological hazard research. In this study, petroleum-exploration data characterising near-surface physical rock properties are re-purposed for a wide range of earthquake-science applications. Here, applications to earthquake-science research include geological and geophysical characterisation of physical rock properties, estimation of site-response parameters used to model shaking intensity, determination of shallow subsurface velocity structure and development of a new generation of high-resolution velocity models required for physics-based simulations of earthquakes and tsunami. Geological and geophysical exploration-well data, derived from lithological and wireline logs, vertical seismic profiles and precise shot timing and navigation from active-source seismic surveys, provide independent constraint and validation for regional velocity models. Relationships between compressional (VP) and shear (VS) wave velocities are investigated for New Zealand’s sedimentary rocks and compared to empirical functions commonly employed for velocity model construction and ground-motion analysis. To a first order, depths to rocks of Mesozoic and older age (shallower than 5 km depth) correlate to VS 2.5 km/s, confirming an important link between key geological and geophysical attributes. VP/VS data from exploration wells are broadly consistent with the current version of the New Zealand-wide velocity model (nzw3.1) and show that empirical relationships between VP, VS and density of sedimentary rocks can be readily improved for local application.Precise active-source travel-time tomographic inversion, supported by MBIE Endeavour, Strategic Science Investment Fund (SSIF) and Royal Society of New Zealand Marsden fund grants, has recently been used to significantly improve the velocity structure of the Hikurangi Margin. The study presented herein updates and extends that velocity model into southern Taranaki Basin and surrounding onshore areas. Precise travel-times from six publicly available industry-standard seismic-reflection surveys have been interpreted from onshore–offshore receiver-gathers generated from 80 GeoNet broadband and short-period seismometers. New and existing basin models from Moutere Basin (near Nelson)and southern Taranaki Basin are incorporated into an initial velocity model prior to precise travel-time inversion. The inclusion of these near-surface (0–10 km) velocity constraints improves characterisation of wave speeds at all depths. In particular, incorporating these independent constraints ensures that travel-time delays associated with the variability in the thickness or wave speed of sedimentary material beneath offshore shot-points are not smeared to greater depth along crustal raypaths. Along with the generation of an updated three-dimensional velocity model for central New Zealand, interpretation of petroleum seismic-exploration data for the depth to VS 1.0 and VS 2.5 km/s are compared with the 2022 New Zealand National Seismic Hazard Model Site Characterisation Database. Initial analysis of seismic-exploration data show an improved correlation between estimates of site period with depth to VS 1.0 km/s for deep basins. Petroleum-exploration seismic-reflection surveys offshore of western North Island (Reinga–Northland and Taranaki basins) and eastern South Island (Canterbury–Great South basins) are available to extend the methodology used in this study to a nationwide scale. Precise travel-time data developed for this and previous studies should be utilised for all future velocity-model development.(auths)
