136
TW. Cranford
and the propagation pathways sum to form what might be termed the "functional signal," along with its manifestations in the resultant interference
patterns that comprise the sound field in the aquatic environment. These
complex acoustic patterns of constructive and destructive interference are
due, in large measure, to the complicated arrangement, topography, and
impedance values of structures within the forehead.
The complexity of the sound field near any real sound source(s) is greater
than it is further away. Consequently, it is easier to characterize sounds in
the far field. In fact, it is common practice to avoid the complexity of the
acoustic near field
3
(White 1991) by recording in the acoustic far field, with
respect to sound source(s). Therefore, by the time most odontocete functional signals are recorded, the focusing and beam-forming effects of refraction, reflection, and interference by the sound propagation pathways upon
the initial signal, within and beyond the animal's head, have already
occurred. Consequently, it is difficult to tease apart the effects of the sound
propagation pathways from the initial signal and the action of the sound
source(s) that produces it.
Some years ago, I saw a series of film loops that may reveal aspects of
the formation of the functional (tissue-borne) signal. Cees Kamminga of
Delft, The Netherlands, constructed a clever demonstration by cinematically overlaying oscilloscope traces of successive pulses in a click train (from
stationary animals). This technique demonstrated, in a graphic way, that
some portions of each (functional) pulse waveform remain stable throughout the click train, while other components are unstable or characterized
by change over time. Specifically, the peaks in the stable portion of each
(functional) pressure waveform maintained a consistent relationship in
the time domain as they grew out of the background and faded back into
it over the course of an entire click train. This characterization was consistent for click trains that contained oligocyclic waveforms (T. truncatus
and Sotafia fiuviatilis), as well as, polycyclic waveforms (P. phocoena and
Cephalorhynchus spp.).
Wiersma (1988) and Kamminga (1981) mathematically demonstrated a
similar separation between stable and variable components of a functional
signal. They showed that if an echolocation waveform (polycyclic or oligo3Dr. Glenn White (1991) defines the acoustic near field: "The sound field very close
to a source of sound is called the near field. By very close is meant less than one
wavelength at the frequency of interest. It is difficult if not impossible to make
meaningful sound pressure level measurements, such as with a sound level meter,
in the near field because the nature of the field itself is very complex. Frequently
the acoustic energy is moving across the surface of the source, or maybe there is a
large air velocity near the source. Standing waves are also present in many cases if
the source is deeply convoluted. In any case, it is not possible to predict the sound
level in the far field from measurements in the near field, so when measuring sound
pressure level, a sound level meter must always be at least one wavelength of the
lowest frequency of interest from the source."
TW. Cranford
and the propagation pathways sum to form what might be termed the "functional signal," along with its manifestations in the resultant interference
patterns that comprise the sound field in the aquatic environment. These
complex acoustic patterns of constructive and destructive interference are
due, in large measure, to the complicated arrangement, topography, and
impedance values of structures within the forehead.
The complexity of the sound field near any real sound source(s) is greater
than it is further away. Consequently, it is easier to characterize sounds in
the far field. In fact, it is common practice to avoid the complexity of the
acoustic near field
3
(White 1991) by recording in the acoustic far field, with
respect to sound source(s). Therefore, by the time most odontocete functional signals are recorded, the focusing and beam-forming effects of refraction, reflection, and interference by the sound propagation pathways upon
the initial signal, within and beyond the animal's head, have already
occurred. Consequently, it is difficult to tease apart the effects of the sound
propagation pathways from the initial signal and the action of the sound
source(s) that produces it.
Some years ago, I saw a series of film loops that may reveal aspects of
the formation of the functional (tissue-borne) signal. Cees Kamminga of
Delft, The Netherlands, constructed a clever demonstration by cinematically overlaying oscilloscope traces of successive pulses in a click train (from
stationary animals). This technique demonstrated, in a graphic way, that
some portions of each (functional) pulse waveform remain stable throughout the click train, while other components are unstable or characterized
by change over time. Specifically, the peaks in the stable portion of each
(functional) pressure waveform maintained a consistent relationship in
the time domain as they grew out of the background and faded back into
it over the course of an entire click train. This characterization was consistent for click trains that contained oligocyclic waveforms (T. truncatus
and Sotafia fiuviatilis), as well as, polycyclic waveforms (P. phocoena and
Cephalorhynchus spp.).
Wiersma (1988) and Kamminga (1981) mathematically demonstrated a
similar separation between stable and variable components of a functional
signal. They showed that if an echolocation waveform (polycyclic or oligo3Dr. Glenn White (1991) defines the acoustic near field: "The sound field very close
to a source of sound is called the near field. By very close is meant less than one
wavelength at the frequency of interest. It is difficult if not impossible to make
meaningful sound pressure level measurements, such as with a sound level meter,
in the near field because the nature of the field itself is very complex. Frequently
the acoustic energy is moving across the surface of the source, or maybe there is a
large air velocity near the source. Standing waves are also present in many cases if
the source is deeply convoluted. In any case, it is not possible to predict the sound
level in the far field from measurements in the near field, so when measuring sound
pressure level, a sound level meter must always be at least one wavelength of the
lowest frequency of interest from the source."