Author Topic: Simulated hazard magnitude  (Read 1212 times)

emongold

  • Newbie
  • *
  • Posts: 28
    • View Profile
Simulated hazard magnitude
« on: February 02, 2022, 02:04:15 AM »
I am wondering if the actual magnitude of the simulated events is available somewhere in the outputs of a set of hazard simulations. For example, if I input to get events from M = 7.0 to M = 8.0 and run 5 scenarios, is there (or could there be) a list somewhere of what magnitudes were chosen for the 5 runs?

kuanshi

  • Jr. Member
  • **
  • Posts: 54
    • View Profile
Re: Simulated hazard magnitude
« Reply #1 on: February 02, 2022, 02:50:06 AM »
Hi emongold,

Yes, the earthquake magnitude and MAF for the ground motions are saved in SiteIM.json, which is located under the local work directory (e.g., "C:/Users/user_name/Documents/R2D/LocalWorkDir/HazardSimulation/GroundMotions/Output").  You can find an attribute "Earthquake_MAF" in the SiteIM.json which reports the earthquake magnitude.  Please feel free to get me back if you have trouble accessing it.

Regards,
Kuanshi

emongold

  • Newbie
  • *
  • Posts: 28
    • View Profile
Re: Simulated hazard magnitude
« Reply #2 on: February 03, 2022, 01:58:28 AM »
Kuanshi,
Thank you- this was helpful! To clarify- there is just one magnitude no matter how many scenarios are run?
Emily

kuanshi

  • Jr. Member
  • **
  • Posts: 54
    • View Profile
Re: Simulated hazard magnitude
« Reply #3 on: February 03, 2022, 07:38:08 PM »
Yes, Emily - the current R2D supports simulating ground motion maps (as defined by the number of ground motion) for the selected scenario in one run - so the magnitude is for that selected scenario.  Please feel free to let us know if you'd like to sample/run multiple scenarios in one job and maybe we could discuss details and upgrade the feature in to the future release.

Best,
Kuanshi

emongold

  • Newbie
  • *
  • Posts: 28
    • View Profile
Re: Simulated hazard magnitude
« Reply #4 on: February 04, 2022, 01:39:32 AM »
Kuanshi,
That makes sense. I would be interested in having R2D be able to run a series with different scenarios for each job. Ideally, a distribution would be useful so when choosing minimum and maximum magnitudes, they have associated frequencies which could reflect the probabilities of various events, but even just a uniform likelihood that each ground motion map comes from some magnitude event within those bounds would be useful.

The way I am using R2D now to simulate ground motions, it seems I need to run scenarios one-by-one and set the bounds from e.g. 7.0 to 7.2 if I want a ~M7.1 event and need to run multiple jobs if I want events reflecting magnitudes between 7.0 and 8.5. Essentially, I am finding liquefaction probabilities and want to know its dependence on the earthquake magnitude along with other parameters, so I want a distribution of magnitudes.

Ultimately, it would be useful to be able to put in a probability distribution so the output ground motions, if running a large enough number, would reflect the likelihood of the events themselves.

Let me know what you think would be possible or if you have other suggestions of how I can use the tool as-is to get these types of outputs more efficiently.
Thanks,
Emily

kuanshi

  • Jr. Member
  • **
  • Posts: 54
    • View Profile
Re: Simulated hazard magnitude
« Reply #5 on: February 07, 2022, 05:13:10 PM »
Hi Emily,

Thank you so much for giving feedback and sharing more details!  Yes, following your suggestions, maybe we can add two new features to the tool: (1) defining the number of earthquake scenarios (this will be in conjunction with the current number of ground motions per site per scenario) and (2) scenario sampling.  The scenario sampling can support (a) using a distribution of magnitude or/and distance (distribution can be uniform or (log)normal) and (b) using the mean annual frequency as the weight.  Please feel free to let me know if these two described features fit to your need (and I would start working on them then), or maybe we could have a follow-up zoom.

Regard,
Kuanshi