New Zealand Ground-Motion Database (Version 4.3)

This dataset contains the 2024 updates to the New Zealand Ground-Motion Database (NZGMDB), originally developed by Hutchinson et al. (See Related Work). A link to the GitHub repository where the previous version resides is available in Related Data and Software. The updated release is designated NZGMDB Version 4.3. The database consists of 6040 earthquakes recorded at 393 stations, with an associated ~70,000 ground motions, and addresses challenges with earthquake source metadata, such as earthquake location, magnitude, tectonic classification, and finite-fault geometry.

Key updates relative to the previous version include:
- Extension of the database to cover seismic events through 2024.
- Expanded earthquake source metadata, finite-fault geometries, and aftershock flagging.
- Enhanced waveform extraction, with longer time windows to better capture significant arrivals.
- Improved P-wave arrival estimates, supported by a comprehensive evaluation of P-wave picker performance—crucial for accurate signal processing.

The quality-screened ground motions are provided in the NZGMDB-Quality-Flatfiles folder, which includes earthquake, station, and record metadata, along with intensity measures for both rotated and orientation-independent components. To provide more flexibility in database use, the repository also includes the NZGMDB-Full-Flatfiles folder. This contains the intensity measures and metadata for the complete set of processed waveforms retrieved from the GeoNet FDSN client, including records that did not comply with the established quality criteria and are absent in the quality flatfiles. Additionally, the raw and processed waveforms are provided in the Waveforms folder, organized by year and GeoNet event identification. Included in this publication are a Readme file that includes a data dictionary about the contents of the files, and a report with summary statistics. Further details on the database are available on the NZ GMDB GitHub Wiki repository including a thorough explanation of the pre- and post-processing techniques, along with the methodologies applied to generate the files provided in this project. The code used to generate the data is also available in the same GitHub repository.