Hello all,
I am currently working on a 6 layers model where some of my layers don't cover the whole area and each layer's thickness is not constant.
To create my grid I use gridgen, this way I am able to refine my first layer as well as to implement my thickness datas for each layers. I then use the DISU module to create my final discretization through
disu_gridprops = g.get_gridprops_disu6()
gridprops_ug = g.get_gridprops_unstructuredgrid()
ugrid = flopy.discretization.UnstructuredGrid(**gridprops_ug)
My issue is that by doing so, the grid of each layer has to cover the whole area and if I put 0 thickness values on the points where the actual layer is non-existent I get messages of error when runing the simulation. For the time being I put thickness values that are almost 0 on those points but this really doesn't seems optimal at all.
Do you have any leads? How do you modelize discontinuous layers while having different grids for each layers? Should I use the grid.add_active_domain() for each layers? (I have not tried it yet because to obtain the actual limits of each layers seems tough and I would rather be sure it works before doing it)
Thanks a lot for your time!