Technical Updates: 2014, Third Quarter

Summary

NGA-East’s latest progress was summarized through a series of presentations at the October 29-30 SSHAC Workshop held at UC Berkeley. Unless otherwise specified, presentations linked below are from the October 2014 SSHAC Workshop. All the presentations for this SSHAC workshop are available here. These presentations often present results from collective work from many participants over extended periods of time. The co-authors list only represent those who prepared the specific presentations.

The project team made considerable progress since the July workshop, notably on the development of Ground Motion Prediction Equations (GMPEs). The October workshop was focused on proponent presentations and discussions of candidate GMPEs, covering both median models and models for the standard deviation of ground motions. The TI-team approach for evaluating the range and space of epistemic uncertainty of the candidate models was also presented. Models presented at the workshop were compared on the basis on the ground motions they predict. As it was the case for past NGA-East SSHAC workshops, scientific progress on various related PEER tasks was also presented at the October workshop.

Data Sets (Task A)

The NGA-East database is the largest, most complete database of uniformly processed ground motions for the Central and Eastern North America region. Various database products have been made available to NGA-East model developers in the last few months. The PEER NGA-East database report was completed and posted online in October. The PEER report is available here.

Updates to database products were presented at the October workshop:

• Database: updates on products and metadata (C. Goulet et al.)
• Database: updates on VS30 assignments (J. Stewart, G. Parker and the Geotechnical Working Group)

Development of Median Ground Motion Prediction Equations (Task H)

Since it’s inception in August 2014, the GMPE Working Group has been meeting weekly to go over various topics of model development. Tremendous progress was made by all participants. The October workshop, which focused on SSHAC workshop element 2 (proponent models and discussions), highlighted the work from these developers. Existing models, such as those included in the EPRI Review Project (2013), had already been summarized at the July workshop and were not presented again this time. Most of the presentations were on new candidate median models (recently developed or under development).

The candidate models were grouped into categories (see slides 8 and 9 here), based on the technical basis for their development. Many models are based, at least in part, on the point-source stochastic model and an overview of the approach was presented by G. Atkinson. This set of presentations was followed by individual proponent model presentations:

• Regional-adjustable generic GMPE based on stochastic point source simulations: application to NGA-East Database (E. Yenier, G. Atkinson)
• SMSIM suite of models (D. Boore)
• Darragh et al. stochastic models (R. Darragh, W. Silva), a model that builds on Kappa estimation from broadband inversion (R. Darragh, W. Silva)
• Reference empirical ground motion model for ENA (B. Hassani and G. Atkinson)
• GMPEs for ENA using the hybrid empirical method (S. Pezeshk, A. Zandieh, K. Campbell, B. Tavakoli)
• GMPE based on broadband synthetic seismograms (A. Frankel)
• Empirical ENA GMPE including intensity and ground motion data (C. Cramer, N. Al Noman)
• Applying modular filter-based approach to CEUS GMPEs modeling (V. Graizer)
• Because it includes a series of new modeling tools, the PEER NGA-East GMPE based on FAS was covered by a set of presentations:
• PEER RVT approach: overview and RVT process, downsampling and smoothing (A. Kottke et al.)
• PEER RVT approach: simulations extrapolation of FAS for M using Finite Fault simulation results (J. Bayless et al.)
• PEER RVT approach: duration model (N. Kuehn et al.)
• PEER RVT approach: empirical FAS model (J. Hollenback et al.)

 
All the candidate models, including those from EPRI (2013), were compared for magnitude and distance scaling:

• Proponent median models – comparison (N. Kuehn for the GMPE Working Group)

 
In addition, available hanging-wall models to be used with the median GMPEs were summarized:

• Summary of NGA-West2 models for hanging wall effects (R. Youngs)

Sigma (Standard Deviation) (Task J)

The Sigma Working Group has developed initial Sigma models for CENA. The summary of existing world-wide models and CENA-specific development was summarized at the workshop:
• General TI Team approach for sigma logic tree; Simulations of PEER NGA-East approach to Sigma, data analysis results and model development and Proponent sigma models (L. Al Atik et al.)

Evaluation of Epistemic Uncertainty (Task K)

The application of the proposed approach (Sammon’s maps) was performed for all the candidate models considered by the NGA-East TI team and presented by (N. Kuehn).
The TI Team approach for median logic tree development and the consideration of epistemic uncertainty (Task K) was summarized (in this presentation from N. Abrahamson).

Regionalization & Source/Path Studies (Tasks C and D)

Teams led by Dr. Mooney (USGS) and Prof. Chapman (Virginia Tech) completed the documentation of their regionalization task. The PEER report is available here.