We might ask if the trees get in the way of amorous vocalization. Vegetation will scatter sound but only if the size of the scattering elements of the sound (trees) is of the same magnitude or larger than the wavelength of the sound. So, consider a large 3 meter tree trunk in a contest with 30 Hz sound with a wavelength of about 15 meters. This Loxodanta African sound is not hindered at all by the forest!
A 15 Hz amorous bellow has a wavelength of 20 meters and the forest is transparent to this sound like sunlight is to glass. Sylvan courting with a deep voice has little loss of efficiency. This topic is covered by Price et al. 1988.
The greatest scattering and least effective communication is for sounds at around 250 Hz. Is it a good idea to issue forth your mating call through a narrow forest gap? This is sort of an ambush model. Carnard-Carauna et al. 1990 tell us that with 63 Hz cooing you gain about 3 dB in this kind of vocal courting.