Ground Motion Selection

Sponsors: Southern California Earthquake Centre (SCEC), University of Canterbury

Collaborators: Peter Powers/Kevin Milner (OpenSHA team), Jack Baker (Stanford)
Postgraduate researchers: Karim Tarbali
Selection of ground motions using the GCIM-based ground motion selection algorithm

Assessment of bias in seismic response due to selected ground motions; and independence of the resulting seismic
demand hazard on the conditioning intensity measure

Ground motion selection provides a critical connection between conventional seismic hazard analysis and timedomain seismic response analyses. Because of the complexity of ground motion time series, in comparison to their
simple characterization, it is often found that there exists significant uncertainty in seismic response as a result of
different incident ground motions. Therefore, if inappropiate/inconsistent ground motion records are selected for use
in seismic response analysis then there is the potential for large biases in response estimates.
This work has focused on the development of a ground motion selection method with a fundamental basis, rather
than heuristic approaches that are widely used. The developed ground motion selection method is based on the
Generalized Conditional Intensity Measure (GCIM) approach (Bradley 2010), which provides the multi-variate
distribution of ground motion intensity measures conditional on a specific value of a single intensity measure (e.g.
SA(T=2.0s)).
A ground motion selection algorithm (Bradley 2012) has been developed to automatically select ground motions
based on the GCIM distributions. The selection algorithm can be used for both as-recorded and simulated ground
motions, and requires only the specification of a 'weight vector', representing a heirarchy of importance of various
intensity measures. A principal feature of the selection algorithm is that ground motions are selected based on their
explicit intensity measure values (e.g. PGA, SA, PGV, Duration, etc), and not based on implicit causal parameters
(e.g. magnitude, source-to-site distance etc) which are utilized in other contemporary ground motion selection
methods. In order to realize the full potential of the GCIM-based ground motion selection algorithm, numerous

intensity measure correlation relationships have been developed (see relevant publications). The ground motion
selection algorithm can also be used to assess any bias in seismic response which may result from inappropiate ground
motion selection.
Because of the theoretical consistency of the GCIM-based ground motion selection algorithm, it has been
demonstrated that the seismic demand hazard can be computed essentially independent on the conditioning intensity
measure (Bradley 2012).
Implementation of the GCIM-based ground motion selection in OpenSHA and wed-based software
In order to facilitate the incorporation of the GCIM-based ground motion selection algorithm and realize its benefits,
it has been incorporated in the open-source, USGS-funded, Open Seismic Hazard Analysis (OpenSHA) software
(See Software and Data, as well as codes for the selection of ground motions) (Scarr and Bradley, 2014).
Ground motion selection for scenario earthquakes
The GCIM-based ground motion selection methodology can also be considered for scenario-based ruptures (Tarbali
and Bradley (2015), Tarbali and Bradley (2014)). Ground motions representative of major earthquake ruptures on the
Alpine, Hope, and Porters Pass faults for Christchurch can be downloaded here; and for major ruptures of the
Wellington, Wairarapa, and Ohariu faults for Wellington can be downloaded here. Full details of this study are
available in a UC report.
Relevant Publications:
1.

Bradley BA, A generalized conditional intensity measure approach and holistic ground motion
selection. Earthquake Engineering and Structural Dynamics 2010, Vol 39 No. 12, pp 1321-1342. (PDF)(link)
2. Bradley BA. A ground motion selection algorithm based on the generalized conditional intensity measure. Soil
Dynamics and Earthquake Engineering Vol 40, pp. 48-61. (PDF) (link)
3. Bradley BA. OpenSHA implementation of the GCIM approach for ground motion selection. 2010 Southern
California Earthquake Centre (SCEC) Annual Meeting. 13-16 September 2010. Palm Springs, California, USA.
(poster). (PDF)
4. Bradley BA. The seismic demand hazard and importance of conditioning intensity measure. Earthquake
Engineering and Structural Dynamics (in press). (link)
5. Bradley BA, Empirical Correlations between Peak Ground Velocity and Spectrum-Based Intensity
Measures. Earthquake Spectra 2012 Vol 28 No 1, pp. 37-54. (PDF) (link)
6. Bradley BA, Empirical correlations between cumulative absolute velocity and amplitude-based ground motion
intensity measures. Earthquake Spectra 2012 Vol 28 No 1, pp. 17-35. (PDF) (link)
7. Bradley BA, Empirical correlation of PGA, spectral accelerations and spectrum intensities from active shallow
crustal earthquakes. Earthquake Engineering and Structural Dynamics 2011 Vol. 40 No. 15, pp 17071721. (PDF) (link)
8. Bradley BA, Correlation of significant duration with amplitude and cumulative intensity measures and its use in
ground motion selection. Journal of Earthquake Engineering 2011 Vol. 15 No. 6, pp 809-832. (PDF) (link)
9. Bradley BA. Displacement spectrum intensity and its use in ground motion selection. Soil Dynamics and
Earthquake Engineering 2011. Vol 31 No 8. pp 1182-1191. (PDF) (link)
10. Scarr J, Bradley BA. An online open-source application for ground motion selection using the GCIM
method. New Zealand Society for Earthquake Engineering Conference. 21-23 March 2014. Auckland, New
Zealand. Paper P3. 8pp.(PDF)