I collected spread spectrum signals at 400 MHz at low elevation angles using Software Defined Radio complex sampling over the seasons. I noticed that over watery, icy surfaces, the RF samples are conjugated compared with regular open space RF samples, resulting in failed de-spreading. De-spreading works if the signal samples are conjugated. I like to think that this phenomenon can be because the 180 deg. reversal of H vector of the wave front in reflection causing |E x H|atan(-H/E) resulting in a 90 deg. change of phase. This answer just doesn't appear to be convincing. Search the Net, could not find any related report on reflected RF signal has signal sample conjugated. Any suggestions?