Assuming these are parabolic antennae, does the inverse square law still apply?
I believe so. A finite parabolic antenna will transmit most of its power into a cone. The cross sectional area of the cone increases with the square of the distance, so the power per unit area (within the cone) falls off as the inverse square.