Modelling Water (liquids)

[ZeidBit]

Hello,

I am modelling a landslide post-earthquake where I initialize the stresses using linear elastic materials then change the GOM file to reflect the residual values of the materials using Mohr coulomb criterion. At the downstream end of my embankment I want to model the runout of water in a canal as shown below:

where Auto41 is the water. When modelling it as a liquid, I always seem to get issues with particles going out of the mesh during the initialization phase or the runout phase. What do you suggest is the best way to model the water moving forward?