When properly computed, the derivatives can then be used to determine the boundary normal. Finally, when both the normal direction and the value of F in a boundary cell are known, a line cutting the cell can be constructed that approximates the interface there. This boundary location can then be used in the setting of boundary conditions.