SEA ICE FRACTURE INVERSION
Sea Ice Mechanics Initiative
Principal Investigator
Prof. Henrik
Schmidt , MIT Dept. Ocean
Engineering
Associate Investigators
Prof. A.B. Baggeroer, MIT Dept. Ocean Engineering
Prof. Ira Dyer, MIT Dept. Ocean Engineering
Prof. J. Robert Fricke
Students
Yuriy Dudko
Qing Wang
Catharina Stamulis
Tarun
Kapoor
Olga Chernets
Collaborating Institutions
Woods Hole Oceanographic Institution
Sponsors
Office of Naval Research
OBJECTIVES
The several Arctic acoustics experiments of the past have
made it clear that ice fracturing is the dominant source of ambient
noise, in turn suggesting that fracturing forms a major component of the
physical processes leading from environmental forcing to the
development of macroscopic ice features such as ridges, leads and
raftings. On this background, the specific objective of the MIT/WHOI
effort under the Sea-Ice Mechanics ARI is to develop a basic understanding
of these ice fracturing
processes and their relation to the environmental forcing, covering
spatial scales of mm to km, and temporal scales of ms to hours. As a
byproduct, the effort is expected to also improve the understanding the
interaction between acoustic waves in the water column and seismic
waves in the ice. Another component of our 1994 effort was aimed at
determining the feasibility of using long range acoustic transmission
to infer changes in global climate. Finally, the large amounts of
acoustic data collected
during the various field experiments is expected to lead to better
understanding of the ambient Arctic acoustic environment. Of
particular current interest is
the marine mammal component. Analysis of our database can provide
unique behavioral data to the marine mammal biologists.
APPROACH
The current
MIT/WHOI effort extends earlier work on acoustic ice fracture inversion,
vastly improving the resolution to allow for direct determination
of source-plane parameters uniquely defining the fracture processes,
such as fracture location, type, orientation, seismic moment and
rupture speed, while at the same time maintaning substantial spatial
coverage. This is accomplished through a coordinated and
consistent effort in experiment design and execution,
seismo-acoustic source and propagation modeling, and advanced signal
processing. In the Spring 94 SIMI experiment, a large
aperture acoustic ``surveillance array'' was used to localize areas of high
seismicity through real-time, adaptive array processing. Such active
areas were then on several occations instrumented with seismic arrays
for monitoring the acoustic emission in the near field, with the
data transmitted to base camp using a
wireless LAN link. The inversion is performed by seismic matched
field processing.
The feasibility of using transarctic acoustic propagation (TAP) to
infer climate changes is addressed by analyzing the SIMI-94 array data
as well. The transmissions from the Russian
source deployed NE of Spitzbergen were received on the horizontal and
vertical hydrophone arrays, and the 6 days of high-quality
transmission data are currently being compared to model predictions.
One of the obstacles to using acoustic thermometry of the oceans is
the potentially harmful effect on marine mammals. To address this
issue, the SIMI-94 data, and data from previous ice camps are being
scanned for marine mammals, and where possible, array processing is
be used to track the whales, reveiling potential behavioral changes
during active source deployments.

Fig. 1. Ice events located by real-time
beamforming on SIMI-94 horizontal hydrophone array between 098:23.00Z and
098:23.59Z, including major events associated with existing crack
immediately N of new shear ridge.
1994 ACCOMPLISHMENTS
A major accomplishment in 1994 was the succesful execution of the
SIMI-94 experiment, in particular the development and implementation
of a new layered approach to the detection of acoustic emission from
natural and artificial ice fractures. A 64-element ``surveillance
array'' was deployed and used for continuous monitoring of the ice
seismicity in a 20 km$^2$ area surrounding the camp.
Figure 1 reproduces the beamformer display
following a major fracture event at the floe boundary N of the array.
As a result of this and other real-time localizations, several deployments
of high-resolution seismic arrays were performed, measuring the
acoustic emission in the near field. In that regard a crucial
accomplishment was the development of the socalled RLAM units
comprised of five 3-component geophones and a hydrophone connected to
a self-contained acquisition system with digital data telemetry to
base camp via a wireless local-area network. This system
allowed for extremely rapid deployment, with a unit being deployed,
the sensors surveyed, and the recording initiated in less than two
hours from the time the deployment was decided.
On the modeling side we have continued the development of the more
complete and accurate seismic source and propagation models needed for
the SIMI fracture plane analysis [Schmidt, 1994]. In particular, we have developed a new
approach for modeling propagation in range-varying elastic waveguides
[Schmidt etal. 1995].
Another significant accomplishment during the SIMI-94 experiment was
the succesful reception of the Transarctic Acoustic Propagation (TAP)
transmissions. The data were of
extremely high quality, allowing for accurate estimation of the modal
arrival structure. In parallel we have refined our Arctic modeling
capability by combining the ice scattering model developed at MIT
[LePage and Schmidt, 1994]
with the KRAKEN coupled-mode propagation model. This combined model has been
succesfully applied to model the TAP data.
1994 APPLICATIONS
The 3-D elastic ice scattering theory developed under this project
has been verified by accurately modeling historical propagation
and reverberation data [LePage and Schmidt 1994]. It has been combined
with the SAFARI/OASES
code to provide a unique capability for waveguide reverberation
modeling [Schmidt 1994]. This model is being applied to ONR sponsored research by
BBN and Alliant Techsystems. The Arctic modal losses predicted by the
model are being used in normal-mode codes for the TAP modeling
performed by SAIC and MIT.
The OASES code with the new 3D seismic source models have been
exported to Lawrence Livermore National Laboratory for use in their
nuclear treaty verification research. In addition updated versions of
OASES has continued to be widely distributed in the ocean acoustic
community in 1994.
REFERENCES
- H. Schmidt, ``Seismo-acoustics of the Arctic ice
cover.'' In L. Bjorno, editor, Underwater Acoustics,
European Commission, Luxembourg, 1994.
- H. Schmidt, W. Seong, and J.T. Goh,
``Spectral super-element approach to range-dependent ocean acoustic
modeling,'' J. Acoust. Soc. Am., In Press: 1995.
- K. Lepage and H. Schmidt, ``Modeling of low
frequency transmission loss in the Central Arctic,'' J. Acoust.
Soc. Am. 96(3), 1783--1795, 1994.