Synopsis
		sfavvvdwe2d < Fwav.rsf vel=Fvel.rsf den=Fden.rsf sou=Fsou.rsf rec=Frec.rsf > Fdat.rsf wfl=Fwfl.rsf verb=n free=n dabc=n adj=n snap=y nb=NOP jsnap=1 dpt=n

The code uses a standard second-order stencil in time.
The coefficients of the spatial stencil are computed
by matching the transfer function of the 6-point discretized
first-derivative operator to the ideal response.

The code implements the linearized operator obtained from the
system of first-order PDEs parametrized in incompressibility and density

dv/dt = - 1./rho * grad(p)
dp/dt = - K * div(v)

where
rho : density
K : incompressibility
div : divergence operator
grad : gradient operator
p,v : pressure and particle velocity wavefields

The models supplied by the user are wave speed and density, the code performs
the conversion internally to buoyancy (inverse density) and incompressibility.

Author: Francesco Perrone
Date: February 2020

Parameters
		bool adj=n [y/n] Adjointness   bool dabc=n [y/n] Absorbing BC   file den= auxiliary input file name   bool dpt=n [y/n] run dot product test   bool free=n [y/n] Free surface   int jsnap=1 undersampling factor for the wavefields   int nb=NOP thickness of the absorbing boundary: NOP is the width of the FD stencil   file rec= auxiliary input file name   bool snap=y [y/n] wavefield snapshots   file sou= auxiliary input file name   file vel= auxiliary input file name   bool verb=n [y/n] Verbosity   file wfl= auxiliary output file name