zero depth trouble

[kate]

With the wet-dry option, it is more likely that someone will attempt to run with a depth of zero somewhere in the grid. The model doesn't put up with that, even if it is inside the land mask. I'm attaching one option for a patch.

[kate]

patch

[arango]

No, this is not a good idea!!! This fix will actually affect the results when there is a lot of beach erosion and will introduce another bug to the code. There is not need to do this because in set_depth there is logic to avoid dividing by zero depth:

# if defined WET_DRY
          IF (h(i,j).eq.0.0_r8) THEN
            h(i,j)=eps
          END IF
# endif

where eps = 1.0E-14_r8 (very small). Recall that this the only place in the code that we divide by depth. Notice that we also take the absolute value of h when taking the squred-root.