Beside an upstanding vertical orrifice, it would be practical to be able to implement horizontal 'bottom holes' from one node to an underlaying node. e.g when a ditch can drain to an underlying conduit by means of a bottom opening. This might be a more straight forward implementation then the weir circumference workround method which is now often used. if possible both circular and rectangular.
Gregory Smith commented
I agree, this is a shortcoming to ICM. Additionally a vertical rectangular orifice would be useful in modelling real world pits.
I will mention that in SWMM software you can have a bottom orifice.