Clarification on Truncated GR vs Incremental MFD Implementation in OpenQuake

[Fatahillah Surya Yudha]

Dear OpenQuake Users,

I hope this message finds you well.

I am currently working on a seismic hazard analysis using a subduction interface source model with the following parameters:

• a-value: 4.271

• b-value: 0.82

• mmin: 6.5

• mmax: 8.5

• bin width: 0.1

In my study, I applied the parameters in two ways (method):

1. Using the Truncated Gutenberg-Richter MFD, inputting the values directly into OpenQuake.

2. Using the Incremental MFD, where I calculated the occurrence rates for each magnitude bin using the formula N=10^(a-b*M), from mmin to mmax with a step size of 0.1.

Unexpectedly, the seismic hazard results from these two methods differ significantly.

However, when I validated the same parameters using nshmp-haz-fortran (from USGS), the results were consistent with method 2 (Incremental MFD).

Is there something I might be missing in the use of the Truncated GR implementation in OpenQuake? Are there specific assumptions or processing steps applied internally that could lead to this difference?

Any clarification or guidance would be greatly appreciated.

Best regards,
Fatahillah Surya Yudha
Sebelas Maret University

[Fatahillah Surya Yudha]

To support this observation, I have attached the resulting hazard maps from all three methods for comparison.  

[Marco Pagani]

The formula you are using for the GR is not correct. See https://github.com/gem/oq-engine/blob/master/openquake/hazardlib/mfd/truncated_gr.py#L102

--
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/718666df-ff3e-4942-9a13-463dfe53e43cn%40googlegroups.com.

[Fatahillah Surya Yudha]

Thank you for your response, Mr. Marco. Allow me to follow up — when I applied the same parameters using the USGS nshmp-haz software, the results were actually quite similar to when i input the occurrence rate with rate of exceedance (M > m).

Could there be a difference in the calculation algorithms between the two software tools? Or are there specific conditions under which we can use the Truncated Gutenberg-Richter model in OpenQuake?

I think I may need further clarification and advice on this. Thank you in advance.

To view this discussion visit https://groups.google.com/d/msgid/openquake-users/98BD7637-6044-4DA4-94E8-5BCBA0E2105D%40globalquakemodel.org.

[Marco Pagani]

Dear Fatahillah, if I recall correctly, the input in the nshmp-haz Fortran code defines the GR in incremental terms.

To view this discussion visit https://groups.google.com/d/msgid/openquake-users/CAPZs5MfBq3NA3dzWtT48E2TMBjTibw8cr6yNoKiK-DDpfzT8gg%40mail.gmail.com.