[CIG-SEISMO] SPECFEM3D how to set magnitude of vertical force

[Dimitri Komatitsch]

Dear Andrea,

In principle, since we do not normalize anything in SPECFEM3D (while in 
SPECFEM3D_GLOBE we switch to non-dimensional values internally), the 
unit of FACTOR_FORCE_SOURCE should be S.I., i.e. Newtons.

However we should just doublecheck that in the source code we do not 
divide that by the Gauss-Lobatto integration weights, the Jacobian or 
anything like that (which would not change the shape of the seismograms 
because the equation is linear, but it would change the magnitude of the 
force and thus the amplitude of the seismogram).

Jo and/or Daniel or Carl, could you confirm that there is no internal 
problem such as dividing by wxgll(), and thus that FACTOR_FORCE_SOURCE 
is in Newtons?

(if so, let us add that to the manual).

Thank you,
Dimitri.

On 10/08/2012 05:23 PM, Andrea Colombi wrote:
> Hello,
> I am trying to use specfem3D for a sort of unusual purpose compared to
> its traditional use. I am replicating small scale laboratory experiments
> about wave-propagation on metal-plate with scatterers and 3D features
> embedded or attached to it. I have been working for a while and the
> results seems overall good. One thing still remain not clear to me: how
> to set the right force magnitude to the source?
> The source I want to emulate acts in the vertical direction
> (perpendicular to the plate, Z-axis) and it applies a load of the order
> of 0.1N. How can i set this value? The code seems to ignore the moment
> tensor parameter when i am using a point force and rather using the
> "FACTOR_FORCE_SOURCE".  I have been following a recent discussion
> (september), which confirmed that by tweaking the code myself i went to
> the right direction, I disabled the original Ricker for the
> Heaviside. Those are the settings I have been using in the constant.h
> file so far:
>
>
> logical, parameter :: USE_FORCE_POINT_SOURCE = .true.
>
> double precision, parameter :: FACTOR_FORCE_SOURCE = 1.d4
>
> integer, parameter :: COMPONENT_FORCE_SOURCE = 3  ! takes direction in
> comp E/N/Z = 1/2/3
>
> and regarding the receivers:
>
> logical, parameter :: EXT_MESH_RECV_NORMAL = .true.
>
>
> I have no UTM projection active and the source is a Dirac pulse (or
> better an Heaviside) as it is originally implemented in the code, i.e.
> reading the netlib error function. Frequency are high (10Khz).
> Thanks for the help and best regards
> Andrea Colombi
>
>
>
>
>
> _______________________________________________
> CIG-SEISMO mailing list
> CIG-SEISMO at geodynamics.org
> http://geodynamics.org/cgi-bin/mailman/listinfo/cig-seismo

-- 
Dimitri Komatitsch
CNRS Research Director (DR CNRS), Laboratory of Mechanics and Acoustics,
UPR 7051, Marseille, France    http://komatitsch.free.fr