Hence, high-frequency signals are attenuated more strongly in closed than in open habitats (Morton, 1975 Marten and Marler, 1977 Wiley and Richards, 1978). However, the slope of the frequency dependence is steeper in dense, forested habitats because of the high degree of sound absorption and scattering from foliage.
Specifically, because of frequency-dependent attenuation, low-frequency sounds transmit generally further than high-frequency sounds. The degree of this degradation depends both on the sound structure and on the acoustic characteristics of the environment (Wiley and Richards, 1982 Brumm and Naguib, 2009). Sounds transmitted through the natural environment are subject to degradation, for example due to environmental absorption, reverberation and scattering. This is known as the acoustic adaptation hypothesis (Rothstein and Fleischer, 1987). Since the 1970s, it has been postulated that the frequency of acoustic signals could reflect an adaptation to maximise the effectiveness of sound transmission in specific environments, such as different habitats (Morton, 1975). Here, we focus on four of the most compelling ones: (1) the acoustic adaptation hypothesis, (2) the morphological constraint hypothesis, (3) the phylogenetic constraint hypothesis and (4) the sexual selection hypothesis. Nevertheless, the frequency of acoustic signals is tremendously diverse across the animal kingdom (Gerhardt, 1994 Fitch, 2006 Gillooly and Ophir, 2010 Pijanowski et al., 2011) and several hypotheses have been proposed to explain this diversity. Low-frequency sounds are generally less attenuated during transmission than high-frequency sounds (Wiley and Richards, 1982 Padgham, 2004). One of the fundamental characteristics of acoustic signals is the frequency of the sound, because it strongly affects signal propagation through the environment (Morton, 1975 Wiley and Richards, 1982 Padgham, 2004). Successful transmission and reception of acoustic signals between conspecifics are essential in diverse contexts, including predation avoidance (alerting others to a threat), territory defence, mate attraction and synchronisation of breeding activities (Bradbury and Vehrencamp, 1998 Catchpole and Slater, 2008). Our results suggest that the global variation in passerine song frequency is mostly driven by natural and sexual selection causing evolutionary shifts in body size rather than by habitat-related selection on sound propagation.Īcoustic signalling is widespread among animals (Bradbury and Vehrencamp, 1998 Gerhardt and Huber, 2002 Catchpole and Slater, 2008).
However, we found no support for the predicted relationship between frequency and habitat.
A phylogenetically informed analysis revealed that song frequency decreases with increasing body mass and with male-biased sexual size dimorphism. Here, we evaluate these hypotheses by analysing variation in peak song frequency across 5,085 passerine species (Passeriformes). Signal frequency may also be under sexual selection because it correlates with body size and lower-frequency sounds are perceived as more intimidating. The acoustic adaptation hypothesis predicts that species in dense habitats emit lower-frequency sounds than those in open areas because low-frequency sounds propagate further in dense vegetation than high-frequency sounds. Animals use acoustic signals for communication, implying that the properties of these signals can be under strong selection.