Classical PSHA-based Damage Calculator - Results

[Liliana Oliveira]

Hi everyone!

I used Classical PSHA-based Damage Calculator to compute damages for a certain region. 

My damage states are: no damage, slight, moderate, extensive and complete.

The output shows a large number of buildings in the state of total damage (above 70% -80% of the total) and almost null values for the state of moderate and extensive damage (~ 2%). 

I think this is strange. I ask if it is really possible to happen or can there be something that I'm doing wrong (?).

Thank you all,

Liliana.

[anirudh.rao]

Hi Liliana,

Would it be possible to share the input files for your calculation at engine....@openquake.org? It will make it easier to answer your question if we can take a look at the specifics of the calculation.

Regards,

Anirudh

[Liliana Oliveira]

I send my input files to engine....@openquake.org.

 

Regards,

Liliana

 

Enviado do Correio para Windows 10

--
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.
For more options, visit https://groups.google.com/d/optout.

 

[anirudh.rao]

Hi Liliana,

Thanks for sharing the input files for the calculation. The issue seems to be with the fragility model used for the calculation — the fragility curves for moderate, extensive, and complete damage states are identical for most of the building types, which is why in your results you see zero values for the moderate and extensive damage states.

Regards,

Anirudh

[Liliana Oliveira]

Hi Anirudh,

 

In an attachment, I send an image with the results that I got from the input files.

I’m so sorry, but I don’t understand why fragility curves influence damage states outputs in that way.

 

Regards,

Liliana.

 

 

[anirudh.rao]

Hi Liliana,

While the fragility functions give the probability of exceeding different damage states at increasing intensity levels, the classical damage calculator gives the number of buildings that are expected to be in (not exceeding) those damage states within the specified time interval. 

Now, if you consider the fragility curves for the four damage states for the building class "BA 1960 1P" in your fragility model, the parameters are the following:

<fragilityFunction id="BA 1960 1P" format="continuous" shape="logncdf">

  <imls imt="PGA" noDamageLimit="0.01" minIML="1e-10" maxIML="5.0"/>

  <params ls="slight" mean="0.2649" stddev="0.92"/>

  <params ls="moderate" mean="0.2742" stddev="0.92"/>

  <params ls="extensive" mean="0.2742" stddev="0.92"/>

  <params ls="complete" mean="0.2742" stddev="0.92"/>

</fragilityFunction>

The parameters for the moderate, extensive, and complete damage states are identical, and those for the slight damage state are nearly the same too. Thus, the curves for the different damage states overlap, as you can see in the figure, and it is no surprise that in your results you see only complete damage or no damage.

Instead, if the fragility functions were to look more like in the second figure below, you could expect to see nonzero values also for the slight, moderate and extensive damage states.

Regards,

Anirudh

[Liliana Oliveira]

Hello,

Thanks for the explanation Anirudh!

So this means that, somehow, complete damage state is representing the other two damage states too, like something cumulative? Or the program just ignore the other two damage states and consider only complete damage states?

Thanks a lot!

Regards,

Liliana

To unsubscribe from this group and stop receiving emails from it, send an email to openquake-users+unsubscribe@googlegroups.com.