I have found a issue in ICM I think explains also my previous posted thread.
Looking at flooding of a 2D manhole, using the graph the flooding on the 2D mesh is null, but looking at the node result data the same flooding onto 2D manhole is 9.5 m3.

I think it's a bug that should be investigated within the software. Or am I missing something?

graph data.pnggrid result.png