[CIG-SEISMO] SPECFEM3D_Cartesian: accuracy in model with water layer

[Youyi Ruan]

Hello,

I have been running a high-resolution test case using SPECFEM3D_Cartesian (V2.0.2) in a local 3-D model includes a water layer and a thin sediment layer. I noticed that in the Par_file of waterlayer_examples, the “OCEANS” is set to “.false.”,  what does the code really do when it set to “.true.” ? (employ the approximation in Komatitsch and Tromp (2002)?) 
 
My simulations are run in a local 3-D model, the element in water layer is about the size of 1km x 1km x 1km.  According to the manual the accuracy of the approximation is about 20 s, this is probably not enough to my case and I am just wondering how can I do to take into account the effects of water phases/effects at period around or shorter than 10 s?  Any suggestions?

Thank you,
Best regards,

Youyi    
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://geodynamics.org/pipermail/cig-seismo/attachments/20130226/3cebfb74/attachment.htm