[CIG-SEISMO] [SW4] difficulties to implement material model

[Petersson, Anders]

Hi Joao,
The input file in examples/rupture/artie-coarse.in runs without problems on my machine. The reason for your difficulties is that you don't specify the material model at the ghost point, which is located just above the free surface. When you use block commands, it is a good idea to put them in order of increasing depth. Furthermore, the first block command should not specify z1=..., and the last should not specify z2=.... That way, the first block command applies to all grid points where z <= z2, and the last applies to all grid points where z >= z1. The intermediate block commands can specify both z1 and z2, even though you would get the same material model if you only specified z1, if the commands are in order of increasing depth. This is because they are parsed from top to bottom, so later block commands supersede earlier ones.  

In your input file, you should first remove the spurious line "absdepth=0", which is not a legal sw4 command. Then, specify the material model with

block vs=450.0 vp=1700.0 rho=2000.0 qs=22.5 qp=45.0
block z1=2.0 vs=650.0 vp=1800.0 rho=2100.0 qs=32.5 qp=65.0
block z1=6.0 vs=850.0 vp=1800.0 rho=2100.0 qs=42.5 qp=85.0
  ...
block z1=21000.0 vp=6800.0 rho=2950.0 qs=190.0 qp=380.0

Hope this helps,
Anders

    ----------------------------------------------------------------------
    
    Message: 1
    Date: Sun, 18 Feb 2018 19:21:10 +0000
    From: João Fontiela <jfontiela at uevora.pt>
    To: cig-seismo at geodynamics.org
    Subject: [CIG-SEISMO] [SW4] difficulties to implement material model
    Message-ID: <2715F897-D7AA-43F7-B3EA-0C01DA951D38 at uevora.pt>
    Content-Type: text/plain; charset="utf-8"
    
    Dear all,
    
    I’m trying implement material model using command block, but SW4 only take in consideration the first line of the block. In consequence the error is the same for all points of computation grid. In the following example I used the block model of example file named artie-coarse.in and I got the same error with my material model:
    
    Point (i,j,k)=(9, 152, -1) in grid g=0
    with (x,y,z)=(5.333333e+02,1.006667e+04,-1.333333e+02) and depth=-1.333333e+02
    is outside the block domain: -1.000000e+05<= x <= 2.000000e+05, -1.000000e+05 <= y <= 2.000000e+05, 2.100000e+04 <= depth <= 8.000000e+04
    
...