The current initialization is given by,
WaveVortexModel(dims, n, z, rhobar, N2, dLnN2, nModes, latitude, rho0)
but the reality is that we don't want the user specifying the z coordinate---those points are the quadrature points, chosen automatically. Also, z=0 will always be the surface, because when we eventually support topography, the depth will be variable. So,
dims [Lx, Ly, Lz], spatial sizes of each domain
n [Nk, Nl, Nj] degrees-of-free/resolved modes
rhobar, N2, dLnN2
latitude, rho0?
I think that'd work. We can then have an option to output any z-grid, but I'm not sure how necessary that is.
In my thinking here, any diagnostic usage of the WaveVortexModel that has some other grid would have to explicitly interpolate to the quadrature grid of the vertical modes. That's fine, and in fact, good.
The current initialization is given by,
WaveVortexModel(dims, n, z, rhobar, N2, dLnN2, nModes, latitude, rho0)but the reality is that we don't want the user specifying the z coordinate---those points are the quadrature points, chosen automatically. Also, z=0 will always be the surface, because when we eventually support topography, the depth will be variable. So,dims[Lx, Ly, Lz], spatial sizes of each domainn[Nk, Nl, Nj] degrees-of-free/resolved modesrhobar,N2,dLnN2latitude,rho0?I think that'd work. We can then have an option to output any z-grid, but I'm not sure how necessary that is.
In my thinking here, any diagnostic usage of the WaveVortexModel that has some other grid would have to explicitly interpolate to the quadrature grid of the vertical modes. That's fine, and in fact, good.