Proposed Conditional Spectrum Calculator & extension to other IMs

[Merrick Taylor]

Hello OQ developers,

I am aware from an earlier post by Michele (topic: mean disaggregation), that there is a conditional spectrum calculator that is in development and will be released as part of 3.13. This is an exciting and important development for ground motion selection in accordance with the Conditional Spectrum approach, thank you for adding this feature.

I have a question that relates to whether it would be possible (it may be relatively easy?) to enable other intensity measures (IMs e.g. PGV, Arias Intensity, Cumulative Absolute Velocity) to be calculated conditionally upon a selected Spectral Acceleration (e.g. PGA, or Sa T=1s etc.).

Currently the new calculator may only be set up for the conditional spectrum. A useful extension would allow other IMs to be calculated that are compatible with the seismic hazard for a chosen spectral acceleration ordinate. In geotechnical engineering we have applications that use PGV, AI, CAV (e.g. seismic pipeline and tunnel design, slope displacement calculation, and liquefaction induced building settlement calculation), and presently it is difficult to obtain this parameter that is accurate and consistent with the hazard specified for building design based on PGA or some Sa ordinate. Correlation coefficients have been published between these alternative IMs and Sa ordinates (happy to share references for some of these) facilitating conditional estimates.

These IMs are usually not included within the same GMPEs for Sa (in some cases PGV may be included, but it is not the norm, and other IMs not at all). I think it would likely require the ability to specify an independent GMM logic tree for the desired IM (such as PGV, AI or CAV), as well as the GMM logic tree for Sa. The underlying calculation of the conditional parameter would otherwise follow the same process as for the conditional spectrum.

I would be interested to hear your thoughts on enabling this in a future update?

Many thanks,

Merrick

[Marco Pagani]

I do not think this will be possible without having an ad-hoc project/collaboration supporting this work (which would require a substantial effort).

While I appreciate that the use of different GMMs for different IML is the best we can do, I am also not convinced that from a methodological standpoint the results would be correct.

Thanks

Marco

Many thanks,

Merrick

--
You received this message because you are subscribed to the Google Groups "OpenQuake Users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to openquake-use...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/openquake-users/4396178c-db34-4023-bf53-3e1223ee3c69n%40googlegroups.com.

MARCO PAGANI | Seismic Hazard Team Lead | Skype mm.pagani | +39-0382-5169863

GLOBAL EARTHQUAKE MODEL | working together to assess risk

GEM - globalquakemodel.org | T - @GEMwrld | F - GEMwrld