Problems with incorrect rendering of ADCIRC meshes with relation to wet/dry cells

[acrosby]

1. Problem with pcolor/filledcontours/contours, where the interpolation required is done over the entire extent of the modeling domain, and not the specific wet cells for that timestep
2. Problem with facet based rendering, where the matplotlib tripcolor technique differs from matlab where a given cell has both wet nodes and dry nodes. All 3 nodes must be wet, for the cell to be considered wet and rendered as a triangular facet.

[acrosby]

The new Matplotlib 1.3.x may solve a few of these problems, and provide the tools to work around the others... need to get some time to try it out.

[dstuebe]

Do FVCOM wet/dry cells work properly?

[acrosby]

I'm not sure, I never found out how FVCOM calculates that. Its likely definitely wrong for contouring/gridded-pcolor styles