Scenario damage using user-defined ground motion fields conditioned on station data and macroseism data

[Valeria]

Dear OpenQuake Community,

I am Valeria, a master's student at UCL (London) who is currently working on her thesis. My research focuses on modeling the scenario damage caused by the earthquake that occurred in Turkey on 02.06.2023, with a magnitude of 7.8 in Kahramanmaras Gaziantep.

Unfortunately, I am encountering some issues with OpenQuake, and I would like to seek further clarification regarding its usage.
Specifically, I am facing difficulties in the modelling phase, especially when attempting to perform a "Scenario calculation by conditioning the ground shaking on station data and macroseism data".

Since in the OpenQuake is mentioned that it is possible to model the damage distribution based on the provided "Ground motion Fields" (page 143 of the manual), I proceeded to calculate the median value of the Ground Motion Field (GMF) (ID132) and used it as input for the scenario damage calculation.

I have proceeded with the modelling of scenario damage (file named "ID138")  using "gmf-data_132." (job file image below) as described in the manual (pag.143) but encountered an error (below).

Here, you can find the folder with the inputs and outputs + the full explanation of the issue.

https://drive.google.com/drive/folders/1WkI6D6fI-13AzzifCxlV_Zkzo15MtzQq?usp=drive_link

I kindly request your help, please. My deadline is at the end of next week, and I would like to complete the scenario damage model by conditioning the ground motion to the stations.

I was wondering if you would be available for a video call to discuss these details more thoroughly and directly. Your assistance would be greatly appreciated.

Thank you for your kindness and expertise.

Best regards,
Valeria.

[Anirudh Rao]

Hi Valeria,

I am unable to replicate the error you are getting. Your damage calculation runs successfully for me. Can you please try to upgrade your engine installation to either v3.16.6-1 or v3.17.2 and check if the error still persists? https://downloads.openquake.org/pkgs/windows/oq-engine/

Best wishes,

Anirudh

[Valeria Pontecorvo]

Hi Anirudh

Thank you so much for you help! It is really appreciated!

So did you run my ID138 folder and it runned perfectly? No errors at all? 

Can you send to me your outputs after damage calculation?

Yeah I will take your advice and I will update it as soon as I can. 

Do you have other versions (v3.17.2)? 

Thank you so much. 

--
You received this message because you are subscribed to a topic in the Google Groups "OpenQuake Users" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/openquake-users/pagmJS3pAyE/unsubscribe.
To unsubscribe from this group and all its topics, send an email to openquake-use...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/openquake-users/8a74510a-c799-48f0-a56e-5dae9b8c5865n%40googlegroups.com.

[Anirudh Rao]

Hi Valeria,

The outputs for ID138 are attached.

Best wishes,

Anirudh

[Valeria Pontecorvo]

Dear Anirudh, 
thank you so much for your help. I installed the v3.17.2 and it runs successfully for me too. I have been stuck for a while on this problem, so I really appreciated your prompt reply.  


Do you confirm that for running the Scenario damage I don't need to use the output called "sigma_epsilon_132.csv"? 

But just "gmf-data_132.csv" and "sitemesh_132.csv"? 

          

Thank you so much again. 
Best wishes.
Valeria Pontecorvo.

To view this discussion on the web visit https://groups.google.com/d/msgid/openquake-users/72ea5eae-99c1-4d52-af66-f112b2f18f91n%40googlegroups.com.

[Anirudh Rao]

Hi Valeria,

You don't need the sigma_epsilon file. Only the gmf and sites files are required.

Best wishes,

Anirudh

[Valeria Pontecorvo]

Hi Anirudh,
I really thank you for your precious help and prompt reply!

Best Wishes,
Valeria Pontecorvo

To view this discussion on the web visit https://groups.google.com/d/msgid/openquake-users/8c0ebb2a-7f08-4f37-97d4-4b35058fb579n%40googlegroups.com.

[Valeria]

Dear OpenQuake Community,

I am reaching out for assistance regarding an issue I am encountering while using the OpenQuake Engine for my research project. I am currently working on assessing the seismic response for buildings with varying numbers of floors, generating specific data sets for each structural configuration.

I have prepared multiple exposure and fragility models for each floor number, along with GMF data derived from seismic hazard analysis. However, while the data sets for floor numbers N10, N9, and N8 have been successfully processed, I am facing a recurring problem with the sets N2, N3, N4, and N7.

The specific error I receive is: ValueError: The sites in gmf_data are disjoint from the site collection. This issue occurs despite following the same data preparation process that worked for the other floor numbers. Additionally, it's worth noting that the data in the folders, except for the exposure model, are the same as those used in a project I worked on last October, (I am working on the same topic), for which I sought your help, and which were processed successfully thank to @Anirudh.

Could you please provide any guidance on how I might resolve or further diagnose this issue?

At the following link, you can download the zip files to be loaded into the machine. These files are related to the number of floors that are processed and others that are not processed, for which I am requesting your assistance.

https://we.tl/t-SpCglMg6Sm

Any help or suggestion would be greatly appreciated, as the proper resolution of this issue is crucial for the progress of my research project.

Thank you in advance for your time and assistance.

Best regards,

Valeria Pontecorvo

[Michele Simionato]

It could be a bug in the GMF exporter not exporting the site ID so that there is not enough information to import it in the following calculation.

It is a tricky thing, it will require a few days to be investigated. I will let you know when it is fixed.

            Michele

[Valeria Pontecorvo]

Thank you so much! 

If you want to I can give you the other data that have been unsuccessful on Openquake. 

I will wait for your reply. 

Thank you again 

To view this discussion on the web visit https://groups.google.com/d/msgid/openquake-users/f348a952-6e60-43a2-8866-16d3f0a00749n%40googlegroups.com.

[Michele Simionato]

On Friday, May 3, 2024 at 4:22:41 PM UTC+2 Valeria wrote:

Thank you so much! 

If you want to I can give you the other data that have been unsuccessful on Openquake. 

I will wait for your reply. 

Thank you again 

It looks like the problem is at the origin, therefore I need the original calculation, the one generating sitemesh_162.csv and gmf-data_162.csv.

Please send the corresponding job.zip.

         Michele

[Valeria]

Hi Michele,

Sorry for the delay in responding, I did not receive a notification of your message. Thank you very much for your help and for your patience. Attached you will find the zip file with the files you requested. Let me know if there is anything else you need. Valeria

[Valeria]

Hi Michele,

I hope this message finds you well.

I am writing to inform you that I have attempted to process my data on OpenQuake again, but unfortunately, without success. Could you please let me know if there are any updates or news regarding the platform that might help me resolve this issue?

Thank you for your support, and I look forward to your response.
Valeria 

[Michele Simionato]

I finally had the time to look at your calculation again. It was REALLY hard to debug. At the end the problem is that the exposure is associated to a site which is on top of a seismic station, which is a case we did not consider, since normally the assets are not  exactly on top of the stations. We will need to come out with a solution, I will let you know when we will have a fix.

                 Michele

[Michele Simionato]

We have now a fix in the current master branch of the engine. Be warned that the conditioned calculator is relatively new and we could change it in the future, as we have done several times already.

       Michele

[Valeria]

Dear Michele,

I want to express my sincere gratitude for the invaluable help you provided in resolving the building data processing issue. Thank you so much for dedicating your time and expertise to resolve the debugging promptly.

 

However, I have found that despite the fix implemented in the current master branch of the engine, I am still encountering the same error while processing my data with both versions 3.17 and OpenQuake Engine 3.20.1-1.

Specifically, the data consistently triggers the same error I was experiencing, namely "ValueError: The sites in gmf_data are disjoint from the site collection!?" in both versions.

I can confirm that other data that was previously processed successfully continues to be processed without issues in both versions.

 

Could you please confirm if you have had the opportunity to process my data using OpenQuake Engine and which version you used, and if the process was successful?

This would help me understand if the issue is specific to my setup.

 

Furthermore, I kindly ask if the problem could be related to the software version, and if so, whether I should consider switching versions or if there are other considerations I should take into account.

I have thoroughly reviewed the release notes for these versions at https://docs.openquake.org/oq-engine/manual/latest/release-notes/whats-new-3.20.html to check for any relevant information regarding my issue.

 

I truly appreciate your ongoing support and am grateful for the assistance you are providing.

Best regards,

Valeria

[Michele Simionato]

You evidently don't know what "master branch" means, don't worry, it is a term used in software development.

In practice it means that the fix has not been released yet, not it has been backported to older releases. You have two options:

1. wait 3-4 months for the next official release 3.21

2. install the master branch now, which is easily done with the universal installer: "python install.py user --version=master"

First, you should remove the version that you have currently installed.

HTH,

           Michele

[Hazem Badreldin]

Sure! Here's a revised version of your message with improved grammar and language:

---

Dear all,

I have run a damage scenario using a fault rupture XML input file, and I would now like to use ground motion fields estimated by another software. I would greatly appreciate it if someone could share an example of the ground motion input file and the job.ini file, so that I can reformat my input files accordingly.

Thank you in advance for your help.

Best regards,
Hazem

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/CADi5W9HvLVdsxdGu3Qgt0kcYVHAhvcYxF0EXw9foyrKsAemmyg%40mail.gmail.com.