[Shake-dev] "Ringing" near intensity IV [SEC=UNCLASSIFIED]

[Bruce Worden]

Hi Trevor,

I think I can reproduce the behavior you've described, but only when using the AW07 IPE. If I use TA09_mmi, or the default IPE (i.e., I leave the IPE unspecified), the MMI seems to behave properly. Can you check to see if you get the same results?

Thanks,
Bruce

On Aug 10, 2011, at 7:37 PM, <Trevor.Allen at ga.gov.au> <Trevor.Allen at ga.gov.au> wrote:

> Hi folks,
>  
> I’ve recently stumbled upon a strange feature in the intensity grids being produced by my ShakeMap installation, and also the installation of colleagues in Indonesia.  We are both using V3.5 - as yet, I have not implemented the latest update Bruce described a month or so ago.  We also both use topo-Vs30 site corrections as is applied in GSM.  Taking this approach, Vs30 values in oceans are typically set to 360 m/s.
>  
> When contouring the mi.grd files, I noticed intensity appears to oscillate about intensity level IV.  At first I thought this was limited to particular GMPE-GMICE combinations since I was originally smoothing the contours.  However, taking a closer look at unsmoothed contours, I noticed the effect was observable (to varying degrees) for all the GMPE-GMICE combinations I tried.  The attached figure shows four scenarios for a hypothetical earthquake of MW 5.7.  As can be observed in all but the bottom-left subplot, the “ringing” about intensity IV is most apparent in oceans where the Vs30 is set to 360 m/s. 
>  
> Hacking into “topo2grd” and modifying the variable “$VS30_OCEAN” to 686 m/s, I reran the scenario and got the bottom-left subplot, in which this ringing effect seems to have disappeared.  Aesthetically, this looks much better, but it does not solve the potential problem for onshore soft-soil sites, which, due to the irregularity of the topo-based approach for determining Vs30, would not be as apparent on the contoured map.
>  
> It’s fair to say that at this point, I’m a little confused as to what might be going on.  I was wondering whether anybody else has encountered this issue?  And if so, is there any easy fix (other than setting $VS30_OCEAN to a higher velocity)?
>  
> Thanks in advance for any advice,
> Trev
>  
> --
> Trevor Allen
> Earthquake Seismologist
> Risk and Impact Analysis & Earth Monitoring Groups
> Geospatial & Earth Monitoring Division
> Geoscience Australia
>  
> T +61 2 6249 9359 | F +61 2 6249 9986
> Physical Address: Cnr Jerrabomberra Ave & Hindmarsh Drv Symonston ACT 2609 Australia
> Postal Address: GPO Box 378 Canberra ACT 2601 Australia
>  
> <contour_plt.unsmoothed.jpg>_______________________________________________
> Shake-dev mailing list
> Shake-dev at geohazards.usgs.gov
> https://geohazards.usgs.gov/mailman/listinfo/shake-dev

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://geohazards.usgs.gov/pipermail/shake-dev/attachments/20110812/7dd1133b/attachment.html>