200
P.L. Tyack and c.w. Clark
This represents a very different kind of sonar from that typically envisioned
by biologists. Echolocation is usually defined as a system in which an animal
listens for echoes of its own sounds. However, sonar engineers distinguish
between a monostatic sonar in which the source and receiver are in the
same place and bistatic sonars in which source and receiver are separated.
Bistatic sonars can either detect sound backscattered from a target, or they
can detect attenuation of the source signal induced by passage through the
target. Each kind of sonar would make specific demands upon auditory processing. Many monostatic sonar processors detect the echo by comparison
to a stored representation of the outgoing signal. In order for an animal to
achieve this, if it was very good at producing the same signal over and over,
then it might compare putative echoes with a permanent representation of
the outgoing signal. If the outgoing signal was variable, the animal might
need to store a temporary representation of each outgoing signal for comparison with putative echoes. If range estimation were important, the auditory system would need to be able to perform precise timing between signal
production and echo return. In order for animals to use a sonar in a bistatic mode, the receiving animal would need to have a clear expectation of
what signal will be emitted by the source animal. In order to achieve this,
either the receiving animal would need to have a precise representation of
the other animal's call stored in long-term memory, or the sending animal
would have to produce a series of the same stereotyped call. In the former
case, the receiver could compare the received call to the stored call, and in
the latter case, the receiver need not have a long-term representation of
the call type, but could simply compare the differences between successive
calls.
Right whales are not known to produce songs, but the call repertoire of
southern right whales, E. australis, on their calving ground, has been exhaustively and quantitatively analyzed by Clark (1982). A principal components
analysis of 10 acoustic features indicated that the vocal repertoire of these
whales formed a continuum. Figure 4.13 illustrates six call types identified
by Clark within this continuum. Four of these call types, up, down, constant,
and high, were tonal, while pulsed sounds were composed of low-frequency
pulses, and hybrid calls had a combination of tonal and pulsive elements.
Clark (1983) compared the rates at which these sounds were produced
as a function of the general activity pattern of the group from which
the sounds were recorded. He divided the repertoire into two functional
groups: a set of discrete calls (up, down, and constant) produced when
whales were swimming, often alone, and a set of highly variable calls (high,
hybrid, and pulsed) produced when whales were in groups that were
actively socializing (Fig. 4.14). The high, hybrid, and pulsed calls were associated with groups producing many visual displays associated with social
interaction, and the more active the groups were, the more of these calls
were produced. The most common of the discrete calls was the up call. This
was most commonly recorded from a whale or group of whales as they were
P.L. Tyack and c.w. Clark
This represents a very different kind of sonar from that typically envisioned
by biologists. Echolocation is usually defined as a system in which an animal
listens for echoes of its own sounds. However, sonar engineers distinguish
between a monostatic sonar in which the source and receiver are in the
same place and bistatic sonars in which source and receiver are separated.
Bistatic sonars can either detect sound backscattered from a target, or they
can detect attenuation of the source signal induced by passage through the
target. Each kind of sonar would make specific demands upon auditory processing. Many monostatic sonar processors detect the echo by comparison
to a stored representation of the outgoing signal. In order for an animal to
achieve this, if it was very good at producing the same signal over and over,
then it might compare putative echoes with a permanent representation of
the outgoing signal. If the outgoing signal was variable, the animal might
need to store a temporary representation of each outgoing signal for comparison with putative echoes. If range estimation were important, the auditory system would need to be able to perform precise timing between signal
production and echo return. In order for animals to use a sonar in a bistatic mode, the receiving animal would need to have a clear expectation of
what signal will be emitted by the source animal. In order to achieve this,
either the receiving animal would need to have a precise representation of
the other animal's call stored in long-term memory, or the sending animal
would have to produce a series of the same stereotyped call. In the former
case, the receiver could compare the received call to the stored call, and in
the latter case, the receiver need not have a long-term representation of
the call type, but could simply compare the differences between successive
calls.
Right whales are not known to produce songs, but the call repertoire of
southern right whales, E. australis, on their calving ground, has been exhaustively and quantitatively analyzed by Clark (1982). A principal components
analysis of 10 acoustic features indicated that the vocal repertoire of these
whales formed a continuum. Figure 4.13 illustrates six call types identified
by Clark within this continuum. Four of these call types, up, down, constant,
and high, were tonal, while pulsed sounds were composed of low-frequency
pulses, and hybrid calls had a combination of tonal and pulsive elements.
Clark (1983) compared the rates at which these sounds were produced
as a function of the general activity pattern of the group from which
the sounds were recorded. He divided the repertoire into two functional
groups: a set of discrete calls (up, down, and constant) produced when
whales were swimming, often alone, and a set of highly variable calls (high,
hybrid, and pulsed) produced when whales were in groups that were
actively socializing (Fig. 4.14). The high, hybrid, and pulsed calls were associated with groups producing many visual displays associated with social
interaction, and the more active the groups were, the more of these calls
were produced. The most common of the discrete calls was the up call. This
was most commonly recorded from a whale or group of whales as they were