Bioacoustics Theories - Soundscape Ecology: Principles, Patterns, Methods and Applications - page 47

Fig. 3.4 Schematic
representation of the effects
of vegetation on sounds
according to frequencies:
high frequencies encounter
more friction in dense
vegetation
In Fig. 3.4 is indicated the excess of attenuation level for random noise and for
pure tone in forest, edge, and grass conditions. This author after playback
experiments has confirmed that forest habitats differ from edge, and open areas in
a narrow range of frequency (1,585-2,500 Hz) have a lower attenuation than lower
or higher frequencies. The attenuation increases sharply above 2,500 Hz. Birds that
inhabit the lower forest level have predominantly pure tones averaging 2,200 Hz.
The edge habitats have a broad range of frequencies, and the attenuation rate
does not differ significantly, suggesting that the selection pressure is not in action in
these habitats. In grassland there is a positive correlation between increasing
frequency and increasing attenuation.
The results presented by Morton ( 1975 ) have been confirmed by Marten and
Marler ( 1977 ) in temperate habitats. Animals vocalizing at an height greater than
1 m can transmit at a longer distance when low frequencies are used. When the
sound is emitted very close to the soil, where the attenuation of lower frequencies is
higher, this allows us to state that the distance at which the sound may be transmit-
ted depends more on the height at which a bird is singing than by the
frequencies used.
Simulations conducted by Brown and Handford ( 1996 ) have confirmed the
acoustic adaptation hypothesis. They constructed two different types of “source
signals,” rapid amplitude modulation (AM) (trills) for open habitats and low AM
(whistles) in close habitats.
The degradation of such signals was performed artificially by various echo
treatments or by various amplitude decrease treatments. The results are consistent
with the AAH because in closest habitats whistled signals transmit with better
quality than trilled signals. In open habitats trilled signals are transmitted with a
lower variability of quality than whistled signals and for their redundancy are
transmitted with more efficiency.
Further experiments conducted by these two authors (Brown and Handford
2000 ) have demonstrated that whistles degrade less than trills in open and forest
habitats but in open habitats trills are received with higher quality than whistles
(Fig. 3.5 ).
The AAH has been proven by Patten et al. ( 2004 ) at the subspecies level in the
song sparrow ( Melospizia melodia heermanni and M. m. fallax ) on the Pacific
slopes of California. These two species occupy riparian habitats, but the first