Page 1 of 1
F-K for active experiments
Posted: Mon Oct 10, 2011 1:39 pm
by lucapuzzilli
Hello dr. Wathelet,
we're working with Geopsy and Dinver to invert some MASW datasets.
I don't know if it depends on my poor skill...but after some comparisons between the "velocity spectrum" obtained with Geopsy plugin and the same spectrum obtained with a commercial software, seems to us that the fk beamforming is able to detect the fundamental mode but seems a little bit poor in detecting 1th and other superior modes. What's your feeling about that?
Thank you
Luca
Re: F-K for active experiments
Posted: Sun Oct 16, 2011 12:55 pm
by admin
Did you try the high resolution option in MASW module?
I'll be back after October 24th. We can further discuss it at that time.
Re: F-K for active experiments
Posted: Tue Oct 25, 2011 8:33 am
by lucapuzzilli
Hello dr. Wathelet,in these few days before you come back to discuss, i prepare some images to better illustrate the comparisons between the velocity spectrums obtained with Geopsy plugin and a commercial software.
http://imageshack.us/photo/my-images/88/shotat20m.png/
http://imageshack.us/photo/my-images/41/shotat25m.png/
To obtain the vel spect with Geopsy we used "Whiten" option, then we processed datasets without using HR , obtaining the images you can see in the pictures (Geopsy images on the right).
Using appropriate palette the Geopsy spectrum visualization could surely enhanced, however we notice that the fundamental modes are more or less well defined, but from 20 to 35 hz FK seems not so efficient to distinguish modes and it's a little bit difficult to distinguish how many modes there are. Thus, in the vel spectrums from Geopsy plugin, when some higher modes "appear"more clear, it's a little bit difficult to pick it along their real """slope""" (compared to the same mode that "appear" more clear on the left of images).
Beyond the differences between the methods implemented to obtain vel spectr (phase shift vs FK beamform) some questions could arise: is always recommended HR option to define vel spectrum just because without using we're not sure that important information could be lost? Using FK plugin without HR could give us some information to better resolve spectrum with HR or it must be always a long session of "trial and error "...?
Let us know if you wish to receive also datasets.
Thank you so much.
Luca
