fault recurrence rate

114 views
Skip to first unread message

Amir jodat

unread,
Nov 29, 2016, 7:15:55 AM11/29/16
to OpenQuake Users
hello every body,

how should I enter fault recurrence rate (for example 0.09) for a simple fault model? 
input files attached!

Im runing OpenQuake 2.1 webui on windows 10 64 bit



with best regards
test-simple CE9.zip

Graeme Weatherill

unread,
Nov 29, 2016, 7:28:26 AM11/29/16
to openqua...@googlegroups.com
Dear Amir,

So there are two options to specify directly the annual rate of recurrence with respect to a given magnitude (or magnitudes). The first is the incremental MFD, the second is the arbitrary MFD. If you are considering just a single magnitude (let's say M 7.0) and its corresponding recurrence rate then the two would work in the same way:

The incremental MFD (note we simply set binWidth to be arbitrarily small - but it cannot be zero!):

<incrementalMFD binWidth="0.001" minMag="7.0">
    <occurRates>0.09</occurRates>
</incrementalMFD>


The arbitary MFD

<arbitraryMFD>
    <magnitudes>7.0</magnitudes>
    <occurRates>0.09</magnitudes>
</arbitraryMFD>


For more than one magnitude the two differ only in terms of the spacing. In the incremental MFD the list of rates is specficied alongside the minimum magnitude and bin width. This means that the resulting MFD will be an evenly spaced pair of magnitudes and rates (with the Mmax implied by the length of the list of rates). The arbitrary MFD is an even more generalised from in which you specify the lists of magnitudes and corresponding rates directly, which in this case need not be evenly spaced.

There is more description and further examples in the manual. Thanks for pointing out the typo, though!
 
Thanks,

Graeme
--
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.


Amir jodat

unread,
Nov 29, 2016, 7:50:18 AM11/29/16
to OpenQuake Users

Dear Graeme

thanks for your reply again. lets explain my issue in more detail. Im trying to verify a paper psha calculation and with next lines info: 

  - Earthquakes of magnitude 5 to 7 were considered and characteristic magnitude-recurrence relationship of Youngs and Coppersmith (1985) was used to model  the probability distribution of magnitudes. 

  - The site is located at a distance 6.7 km from the fault with Vs30 = 760 m/sec2

  - fault is assumed to have a recurrence rate of 0.09 earthquakes per year. 

  - Rupture lengths of earthquakes were a function of magnitude, as determined using Wells and Coppersmith (1994).
  - 
 the 2% in 50 year uniform hazard spectrum from PSHA analysis is shown

 

as you mentioned I have to calculate annual rate of recurrence manually and enter them by arbitraryMFD for all magnitudes, am I right?


Amir

 


Graeme Weatherill

unread,
Nov 29, 2016, 8:12:14 AM11/29/16
to openqua...@googlegroups.com
Dear Amir,

Without knowing more about the exact process the authors are doing it is difficult to say for certain. My guess from the description is that they are using the Youngs & Coppersmith (1985) MFD anchored to the rate of the characteristic magnitude. Note that "characteristic magnitude" in this case is less than the maximum magnitude (usually by 0.25 magnitude units assuming the 0.5 unit characteristic bin width suggested by Youngs & Coppersmith). So it is ambiguous from that description as to whether the magnitude 7 refers to the maximum magnitude considered in the MFD (i.e. characteristic magnitude = 6.75), or the characteristic magnitude itself (i.e. maximum magnitude is 7.25). For this purpose I will assume M 7 is the characteristic magnitude.

OpenQuake can define the Youngs and Coppersmith from either a combination of characteristic magnitude and characteristic rate, or from characteristic magnitude and total moment rate. Examples of both can be seen here: https://github.com/gem/oq-engine/blob/master/openquake/commonlib/nrml_examples/source_model/alternative-mfds.xml. In your case you need the former, which should look like this (assuming a b-value of 0.8 and a bin width of 0.1):

<YoungsCoppersmithMFD bValue="0.8" binWidth="0.1" characteristicMag="7.0" characteristicRate="0.09" minMag="5.0"/>

For the fault and the site you will need to configure the source and site files accordingly.

Thanks,

Graeme
--

Amir jodat

unread,
Nov 29, 2016, 11:33:44 AM11/29/16
to OpenQuake Users
Thanks Graeme, it was helpful


On Tuesday, 29 November 2016 15:45:55 UTC+3:30, Amir jodat wrote:
Reply all
Reply to author
Forward
0 new messages