420
J.L. Aroyan et al.
data. The vestibular sacs, however, were only marginally discernable. By
extracting all CT data coordinates with values below soft tissue thresholds,
the narial sacs and portions of the vestibular sacs were located. Unfortunately, the inferior vestibule, the spiracular cavity, and the premaxillary,
nasofrontal, and accessory sacs were not even marginally discernible in this
data. Nevertheless, the location of the premaxillary sacs is so well documented in the literature that the author deemed it permissible to include
their positions above the premaxillary shelves of the skull in the model
(Mead 1972; Dormer 1979). Note that portions of both left and right
vestibular sacs, as well as other sacs in their entirety, are missing from this
model.
As in previously reported simulations (Aroyan 1990; Aroyan et al. 1992),
air sacs were simulated as pressure release surfaces by setting the pressure
to zero at all sac coordinates. Note that this procedure eliminates any potential air-cavity resonance behavior.
3. Results of the Simulations
To provide an overview of the acoustical focal behavior of the head of the
common dolphin, consider first the result of ensonifying a full head model
with a (cw) 50-kHz sound beam incident from directly forward of the
animal. Figure lOAa illustrates the skin isosurface of this full model, which
incorportated the skull, soft tissues, upper nasal air sacs, the peribullar
sinuses surrounding most of the inner ears, and the pterygoid sinuses of the
lower skull and palate. The full head model utilized the entire CT data set
at half resolution (mapped to a 3.0-mm cubic grid). Figure lOAb illustrates
an isosurface at 30% of the maximum total acoustic energy density Wtotal
(Eq. (2)) within the same model volume (and perspective) illustrated in
Figure lOAa. Significantly, three bright focal points occur within the model,
each with an associated "funnel" or "channel" of energy density leading up
to it. Funnels occur within both left and right fat bodies of the lower jaw,
which narrOW back to create sharp maxima against the lateral surfaces of
each respective (left and right) tympanoperiotic complex. This channeling
behavior of the lower jaw fat bodies strongly improves the forward receptivity of the simulated hearing patterns of both left and right ears (Aroyan
1996). The thinner-at-the-center thickness profile of each (left and right)
lower jaw pan bone surrounded by low-speed fat apparently acts as a "fast"
lens structure contributing to the creation of distinct forward receptivity
peaks for each ear. The peribullar sinuses were also found to be highly effective in shielding the ears from directions of incidence other than forward
and downward. All three results suggest mechanisms potentially fundamental to lower jaw hearing by delphinid cetaceans (Norris 1968, 1980);
however, project results concerning the biosonar reception process must
J.L. Aroyan et al.
data. The vestibular sacs, however, were only marginally discernable. By
extracting all CT data coordinates with values below soft tissue thresholds,
the narial sacs and portions of the vestibular sacs were located. Unfortunately, the inferior vestibule, the spiracular cavity, and the premaxillary,
nasofrontal, and accessory sacs were not even marginally discernible in this
data. Nevertheless, the location of the premaxillary sacs is so well documented in the literature that the author deemed it permissible to include
their positions above the premaxillary shelves of the skull in the model
(Mead 1972; Dormer 1979). Note that portions of both left and right
vestibular sacs, as well as other sacs in their entirety, are missing from this
model.
As in previously reported simulations (Aroyan 1990; Aroyan et al. 1992),
air sacs were simulated as pressure release surfaces by setting the pressure
to zero at all sac coordinates. Note that this procedure eliminates any potential air-cavity resonance behavior.
3. Results of the Simulations
To provide an overview of the acoustical focal behavior of the head of the
common dolphin, consider first the result of ensonifying a full head model
with a (cw) 50-kHz sound beam incident from directly forward of the
animal. Figure lOAa illustrates the skin isosurface of this full model, which
incorportated the skull, soft tissues, upper nasal air sacs, the peribullar
sinuses surrounding most of the inner ears, and the pterygoid sinuses of the
lower skull and palate. The full head model utilized the entire CT data set
at half resolution (mapped to a 3.0-mm cubic grid). Figure lOAb illustrates
an isosurface at 30% of the maximum total acoustic energy density Wtotal
(Eq. (2)) within the same model volume (and perspective) illustrated in
Figure lOAa. Significantly, three bright focal points occur within the model,
each with an associated "funnel" or "channel" of energy density leading up
to it. Funnels occur within both left and right fat bodies of the lower jaw,
which narrOW back to create sharp maxima against the lateral surfaces of
each respective (left and right) tympanoperiotic complex. This channeling
behavior of the lower jaw fat bodies strongly improves the forward receptivity of the simulated hearing patterns of both left and right ears (Aroyan
1996). The thinner-at-the-center thickness profile of each (left and right)
lower jaw pan bone surrounded by low-speed fat apparently acts as a "fast"
lens structure contributing to the creation of distinct forward receptivity
peaks for each ear. The peribullar sinuses were also found to be highly effective in shielding the ears from directions of incidence other than forward
and downward. All three results suggest mechanisms potentially fundamental to lower jaw hearing by delphinid cetaceans (Norris 1968, 1980);
however, project results concerning the biosonar reception process must