Re: The Conditional Mean Spectrum (CMS)

[Marco Pagani]

It’s impossible to answer without seeing the input.

Hi All

I am trying to run for CMS and I get this result why? Anybody has any idea?

Thanks

--
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 visit https://groups.google.com/d/msgid/openquake-users/ecbb6af2-4a6a-4440-b594-7844e8f65091n%40googlegroups.com.

[Marco Pagani]

The only thing I am spotting in the .ini file is that you have an investigation time of 50 years, and a small probability of exceedance

Hi Marco

Please see the attached 

Thanks

To view this discussion visit https://groups.google.com/d/msgid/openquake-users/c278fa65-98d0-4cfd-b4b4-b3752a750436n%40googlegroups.com.

[Marco Pagani]

Thanks. I am currently too busy to work on this. Please try to adjust your ini file to have a more sensible probability of exceedance.

Hi Maro

Please see the inputs I use for the modelling, there are two source models, other one is more than 24MB and I cannot upload here, it is a smooth model

and both source models include three different tectonic settings, 

To view this discussion visit https://groups.google.com/d/msgid/openquake-users/eabf9eb8-ed5a-41be-9e21-141f9a443860n%40googlegroups.com.

[Hiwa ATC]

Hi Marco

  Could you clarify what is wrong with the POE? I need to generate the CMS from Uniform Hazard Spectrum (UHS) for a 10,000-year return period . 

investigation_time = 50.0 and poe=0.005

thanks

[Hiwa ATC]

anybody could help me with this issue? I reckon the issue maybe CMS can be done in the engine when we are using Smooth Seismic Source Model

am i right?

[Marco Pagani]

I suggest you try to run an analysis for a higher probability of exceedance.

To view this discussion visit https://groups.google.com/d/msgid/openquake-users/24551d8a-59e2-4fea-9906-43dd02a92bb1n%40googlegroups.com.

[Hiwa ATC]

Thanks Marco

I have developed UHS for a Dam structure and its CCA requires to do UHS for 10,000 year return period. any other idea? that i can do that?

[Marco Pagani]

Since you are having problems, I suggest trying shorter RPs. Note that, ideally, for a dam, it would be advisable to incorporate potentially nearby structures.

To view this discussion visit https://groups.google.com/d/msgid/openquake-users/2f64f0ef-c045-4ec6-b337-45a2b872db3an%40googlegroups.com.

[Hiwa ATC]

  There are no identified structures close to the site. The UHS has been developed based on a PSHA for a 10,000-year return period, PGA is approximately 1.5g.

I think it may be a bug in the engine when we are using smooth model. 

Thanks  

[Hiwa ATC]

Hi All

  Has anyone encountered this issue before?

Thanks

[kendra....@globalquakemodel.org]

Hi, 

It's not possible to run your job (even changing the logic tree in the ssmLT to exclude the too large file) because the TRTs included in the source model don't match those in the gmmLT. Without the complete job, or one that exactly reproduces the possible bug, we can not investigate.

The conditional spectrum calculator is a post-processing function so it is not aware of the source typologies you have used, i.e. the function does not know the difference between a smoothed seismicity model (presumably you mean point sources or multipoint sources) or another typology.

Did you check that the values in the computed UHS are logical? 

[Hiwa ATC]

Hi 

I wasn’t able to share the real data as it contains confidential information, so I’ve uploaded sample source models instead.

Would it be possible to have your email address so I can send you the actual input files directly? The computed UHS looks logical and makes sense, but I’d like to perform a Conditional Mean Spectrum (CMS) analysis as well.

Thanks

[Anirudh Rao]

As mentioned in the forum description, any files that cannot be publicly disclosed but are needed for debugging can be sent to engine....@openquake.org.