Quarterdeck 3.3
Tiny Bubbles
by Tony Lyons
Scientists have long known that small bubbles of air have
a strong effect on the propagation of sound waves in the ocean, largely
due to the different physical properties of gases and liquids. Seafloor
sediments can also contain gas bubbles which are usually filled with methane.
The gas bubbles can absorb energy out of an incoming sound wave and redirect
or scatter the energy in all directions. These bubbles have an impact on
acoustic remote sensing systems, such as subbottom profilers and sidescan
sonars, and subsequently affect our ability to identify and characterize
the properties of the seafloor effectively. Part of my dis-sertation focused
on gaining a better understanding of the effect of scattering of acoustic
waves by small bubbles in seafloor sediments. I used acoustic-scattering
computer models, seafloor sediment samples, and acoustic data.
[104K] This vertical section image
reconstructed from CT scan data shows methane bubbles inside the sediment
of an Eckernforde Bay core. The gas content (volume concentration %) is
directly related to these bubbles.
My study was greatly enhanced by an experiment carried out in the spring
of 1993 by the Naval Research Laboratory (NRL) as part of their Coastal
Benthic Boundary Layer Program. This experiment took place in Eckernförde
Bay, Germany on the Baltic Sea. In many parts of Eckernförde Bay biogeochemical
processes (bacterial activity) cause layers of methane bubbles to form in
the seafloor. These layers significantly affected the response of acoustic
remote-sensing systems that were in use during the experiment. Sediment
cores were taken in locations surveyed using NRL's Acoustic Sediment Classification
System (ASCS). The colocation of cores and acoustic data allowed the perfect
chance to use the core information in a bubble scattering model which could
then be compared with the ASCS data.
To prevent gas in the bubbles from expanding as samples were raised, cores
from Eckernförde Bay were sealed in pressure-tight containers on the
seafloor then transferred to the surface where they were scanned using x-ray
computed tomography (CT or CAT scans) while still under pressure. The CT-scan
images of the cores showed the layers of bubbles as they existed when the
sediment was in place on the seafloor. With information from the CT scans,
the sizes, depths, and number of bubbles were determined. The bubbles measured
between 0.5 millimeters (the smallest bubble a CT scan can discern) to 8.0
millimeters in radius. Most of the bubbles seen in the CT scans, especially
the larger ones, were not spherical but coin shaped. These "coins"
of gas were aligned vertically, rather than horizontally as expected, an
attribute which remains unexplained.
The computer model I used to analyze the scattering of acoustic signals
by bubbles differed in several aspects from models used to study bubbles
in water. My model had to account for the stiffness that the sediment frame
provided for the bubbles as well as their non-spherical shape. Using the
model I discovered that a non-spherical bubble can have a larger scattering
effect on acoustic signals than a spherical bubble of the same volume. Thus,
a non-spherical bubble in sediment will appear to have larger volume than
it actually has when surveyed acoustically. When included in an algorithm
to account for contributions from all bubbles in the layers, model predictions
of the return of acoustic signals agreed with ASCS data. Both showed long
returns (in time) from the bubble layers and high attenuation within the
layers. The acoustic response of the bubble layers over a range of fre-quencies
calculated with the model showed that higher fre-quencies might penetrate
a sediment better when its bubble distribution resembles that of Eckernförde
Bay.
The study of scattering by bubbles in gassy sediments involved the interplay
of theoretical and experimental work. In this case it was possible to construct
a model of the seafloor which captured enough of the complexity of the real
seafloor to simulate observed acoustic returns. Thus validated, the model
is more open to testing. In this way greater insight into the interaction
of sound with gassy sediments can be obtained. This can lead to better ways
of remotely sensing sediments which contain gas bubbles or better ways to
remotely sense the bubbles themselves if they are the object of interest.
Editor's Note: Tony graduated from Texas A&M in May 1995
with a Ph.D. He works as a scientist for NATO's SACLANT Undersea Research
Centre in La Spezia, Italy, where he continues to study the interaction
of sound and the ocean environment. He can be reached at lyons@saclantc.nato.int.
[Next | Previous
| Contents | Home]
Oceanography, Texas A&M
University
rshatto@ocean.tamu.edu
URL=http://oceanography.tamu.edu/Quarterdeck/QD3.3/Lyons/lyons.html
Updated January 8, 1996
[104K] This vertical section image
reconstructed from CT scan data shows methane bubbles inside the sediment
of an Eckernforde Bay core. The gas content (volume concentration %) is
directly related to these bubbles.