quencies for a linearly swept, sinusoidal signal (100 Hz–20 kHz) in a very
shallow (5–45 cm depth) freshwater pond with a layered substrate of soft
sediment (clay silt and leaf litter) over hard, packed clay. Their results
support the prediction that the effect of substrate on cutoff frequency varies
with depth, acting more like a rigid or soft bottom with decreasing or increasing depth, respectively. Given a constant depth and no temperature
gradient, the wavelength of the cutoff frequency was predicted to lie between two and four times the depth of the water as predicted for soft and
rigid bottoms, respectively.
10.3. Geometrical Spreading
It has been proposed that the propagation of the sound wave in shallow
water might be best modeled as cylindrical spreading because the air–water
interface and the bottom act as boundaries that limit sound radiation (Urick
1983). In this case, pressure is expected to decrease by 10 * log 10 (r) rather
than 20 * log 10 (r), which is associated with spherical spreading (see Sections
4.1.1 and 4.1.2). Banner (1970) used pressure-sensitive hydrophones to map
the sound field of pulsed and continuous noise signals in a very shallow
(20–60 cm) bay with a flat, sand-mud bottom. Pressure loss was 8–12 dB per
distance doubling over the 80–320-Hz frequency band for a continuous
random noise, clearly exceeding the predictions for either spherical or cylindrical spreading losses; transmission loss (TL) in this case was about 25
to 40 * log 10 (r). By contrast, as predicted for cylindrical spreading, pressure
loss was 3 dB per distance doubling for pulsed signals such as the splashes
from leaping fish, chewing or grating sounds from feeding fish and crustaceans, and swimming-associated sounds represented by 0.01–1.2 sec pulses
of broadband noise. Given the very shallow depths for this study, TL values
were likely closer to theoretical predictions because of their higherfrequency content. Other studies of transmission loss associated with the
propagation of natural sounds will be discussed in a later section.
11. Baleen Whale Acoustics
As discussed above, low-frequency sound in deep, temperate water is
refracted into a sound channel where transmission loss is on the order of
10 * log 10 (r). Payne and Webb (1971) hypothesized that the loud, infrasonic
sounds of certain balaenopterid whales are adapted to take advantage of
this physical acoustic niche. They further postulated that, prior to modern
shipping, whales might have communicated across ocean basins (see also
Patterson and Hamilton 1964; Norris 1966). Here, we will further explore
this idea through a simple approach of comparing extant species that
inhabit either shallow or deep water habitats. Is there a consistent relationship between the general acoustic characteristics for different whales
2. Physical Acoustics of Underwater Sound Communication
35
shallow (5–45 cm depth) freshwater pond with a layered substrate of soft
sediment (clay silt and leaf litter) over hard, packed clay. Their results
support the prediction that the effect of substrate on cutoff frequency varies
with depth, acting more like a rigid or soft bottom with decreasing or increasing depth, respectively. Given a constant depth and no temperature
gradient, the wavelength of the cutoff frequency was predicted to lie between two and four times the depth of the water as predicted for soft and
rigid bottoms, respectively.
10.3. Geometrical Spreading
It has been proposed that the propagation of the sound wave in shallow
water might be best modeled as cylindrical spreading because the air–water
interface and the bottom act as boundaries that limit sound radiation (Urick
1983). In this case, pressure is expected to decrease by 10 * log 10 (r) rather
than 20 * log 10 (r), which is associated with spherical spreading (see Sections
4.1.1 and 4.1.2). Banner (1970) used pressure-sensitive hydrophones to map
the sound field of pulsed and continuous noise signals in a very shallow
(20–60 cm) bay with a flat, sand-mud bottom. Pressure loss was 8–12 dB per
distance doubling over the 80–320-Hz frequency band for a continuous
random noise, clearly exceeding the predictions for either spherical or cylindrical spreading losses; transmission loss (TL) in this case was about 25
to 40 * log 10 (r). By contrast, as predicted for cylindrical spreading, pressure
loss was 3 dB per distance doubling for pulsed signals such as the splashes
from leaping fish, chewing or grating sounds from feeding fish and crustaceans, and swimming-associated sounds represented by 0.01–1.2 sec pulses
of broadband noise. Given the very shallow depths for this study, TL values
were likely closer to theoretical predictions because of their higherfrequency content. Other studies of transmission loss associated with the
propagation of natural sounds will be discussed in a later section.
11. Baleen Whale Acoustics
As discussed above, low-frequency sound in deep, temperate water is
refracted into a sound channel where transmission loss is on the order of
10 * log 10 (r). Payne and Webb (1971) hypothesized that the loud, infrasonic
sounds of certain balaenopterid whales are adapted to take advantage of
this physical acoustic niche. They further postulated that, prior to modern
shipping, whales might have communicated across ocean basins (see also
Patterson and Hamilton 1964; Norris 1966). Here, we will further explore
this idea through a simple approach of comparing extant species that
inhabit either shallow or deep water habitats. Is there a consistent relationship between the general acoustic characteristics for different whales
2. Physical Acoustics of Underwater Sound Communication
35