CRREL Stress Sensor Data
Coordinated Eastern Arctic Experiment (CEAREX)
W.B. Tucker, III and D.K. Perovich
U.S. Army Cold Regions Research and Engineering Laboratory (CRREL)
72 Lyme Road
Hanover, NH 03755-1290 USA
December 1990
This paper discusses a stress measurement program in which
biaxial vibrating wire stress sensors were used to monitor pack
ice stresses on the drift phase of the Coordinated Eastern Arctic
Experiment (CEAREX). On the drift phase, the Norwegian charter
vessel Polarbjorn was moored to a multiyear floe and allowed to
drift with the pack while the behavior of the ice, ocean and
atmosphere were monitored. After initial emplacement in the ice
with the assistance of the U.S. Coast Guard Icebreaker Northwind,
the Polarbjorn drifted from 82 deg 40 minN, 32 deg 26 min E to
80 deg 10 min N, 31 deg 13 min E from 18 September to 25 November 1988.
This joint experiment provided an ideal setting for
measuring ice stresses because they could be correlated with
other environmental parameters. The primary objective of the
stress program was to monitor pack ice geophysical stresses in
the eastern Arctic for the two month drift period. It was
anticipated that the ice would drift around the north coast of
Spitsbergen and into the Fram Strait. We believed that large ice
stresses would be experienced as the ice was compressed against
Spitsbergen as it entered Fram Strait. Instead, the local pack
ice drifted to the east of Spitsbergen and entered the Barents
Sea. Nonetheless, large stresses were observed during
deformation of the ice as it compressed and sheared against
Kvitoya Island. Another major objective was to investigate
vertical and horizontal variations in stress on the floe. This
was facilitated by the use of the vibrating wire sensors which
are easily installed at any depth. A final objective was to
compare stresses in thin first-year ice to those in adjacent
multiyear ice.
High quality pack ice stress data was obtained during the
CEAREX drift program. We were fortunate to monitor stresses during the
complete disintegration of floes in the area when the large
deformation at the end of the experiment occurred. Data from
different vertical levels and different horizontal locations were
obtained.
FIELD MEASUREMENTS
Stress Sensors
The stress measurements described in this investigation were
obtained with biaxial vibrating wire stress sensors similar to
those described by Cox and Johnson (1983). These instruments were
chosen because of their relative ease of installation and
maintainence, the extensive calibrations that had been previously
performed (Cox and Johnson, 1983), and their success in earlier
field programs Johnson et al, (1985). The sensor consists of a
stiff steel cylinder which is 0.25 m long, 57 mm in diameter and
has a wall thickness of 16 mm. The gauge is much stiffer that
the ice, having a modulus of about 200 GPa, thus should not be
affected by spatial variations or temporal changes of the ice
modulus. Six tensioned wires are set at 30ø intervals across the
inside diameter of the gauge. Only three wires are necessary for
the determination of principal stress components; the remaining
wires are for redundancy.
Field Installation
The Polarbjorn was moored to a multiyear floe on September
16, 1988 at 82 deg 40 min N, 32 deg 26 min E. Four stress sensor
sites were established over the next two weeks, on each of two adjacent
multiyear floes. Figure 2 shows the configuration of the stress
sensor sites on the two floes. At site 1, on floe Alpha, three
sensors were installed at vertical levels near the top, mid-depth
and bottom of the ice sheet. At site 2, which was located on the
edge of the multiyear floe Beta adjacent to a freezing lead, one
sensor was placed in the first-year ice while the remaining two
were located at shallow depths in the multiyear ice. Because of
sensor and data logger difficulties, the data from sites 3 and 4
were generally unreliable and are not reported on here.
The sensors were installed in holes approximately 0.1 m in
diameter made with a standard ice coring auger. Each sensor is
attached to a length of PVC pipe which is suspended to the proper
depth from the surface by a cross bar through the PVC extension.
Once the sensor was placed in the hole at the proper depth, and
leveled in the horizontal plane, the hole was filled with fresh
water. Depending upon the air and ice temperature, the hole was
normally completely frozen within a day or two. Stresses
associated with the freeze-in can be 200-300 kPa, and usually
take several days to dissipate.
At site 1, the multiyear ice was 1.60 m thick. Sensors were
installed at depths of 0.25, 0.70 and 1.20 m. At site 2, one
sensor was located in the first-year ice at a depth of 0.20 m.
The thickness of this frozen lead increased from 0.38 to 0.54 m
during the stress monitoring period. The sensor was located
about 7.0 m from the edge of the multiyear floe in the thin ice
which was about 200 m wide. The two remaining sensors were
located in 2.0 m thick multiyear ice, at distances of 2 and 15 m
from the edge of the floe. The three sensors were configured in
a straight line normal to the edge of the floe and the frozen
lead.
Data from each site was recorded on Campbell Scientific,
Inc. data loggers. The sensors were sampled and recorded at two
minute intervals, and the period (frequency) of each wire in each
sensor was stored. The data loggers cached data in memory
modules which required replacement with empty modules at 5 to 7
day intervals. Normally, no interruption in sampling occurred
during this exchange, though there were occassional breaks in the
data if the memory modules were filled to capacity.
Data Set Description
The CRREL stress sensor dataset consists of time series of stress
invariants as determined from vibrating wire stress sensors. There are six
files, each containing the entire time series of data from one sensor.
A file consists of a header line followed by data, with each data
line containing
a) time in decimal Julian day;
b) the first stress invariant;
c) the second stress invariant.
The format is comma-separated variables. Missing or bad data is coded
as '******'.
Filename Sensor Ice Distance From Ice
Depth Thickness Floe Edge Type
========== ====== ========= ============= ==========
Group 1
CRREL1A.INV 0.25 m 1.6 m 200 m Multi-year
CRREL1B.INV 0.70 m 1.6 m 200 m Multi-year
CRREL1C.INV 1.20 m 1.6 m 200 m Multi-year
Group 2
CRREL2A.INV 0.20 m 0.6 m 7 m Young ice
CRREL2B.INV 0.25 m 2.0 m 2 m Multi-year
CRREL2C.INV 0.25 m 2.0 m 15 m Multi-year
References
Cox, G.F.N. and J.B. Johnson (1983). "Stress Measurements in Ice,"
U.S. Army CRREL Report 83-23, 31 pp.
Johnson, J.B., G.F.N. Cox and W.B. Tucker III (1985) Kadluk ice stress
measurement program. In: International Conference on Port and Ocean
Engineering under Arctic Conditions (POAC), 8th, Narssarssuaq, Greenland.
Proceedings, vol. 1, pp. 88-100.
Tucker, W.B., D.K. Perovich, M.A. Hopkins, and W.D. Hibler III.
"On the Relationship Between Local Stresses and Strains in Arctic Pack Ice,"
Annals of Glaciology 15 (August 1990): 26-31. Proceeding of the Symposium on
Ice-Ocean Dynamics and Mechanics held in Hanover, New Hampshire, U.S.A.
Tucker, W.B. III and D.K. Perovich. "Stress Measurements in Drifting
Pack Ice," Cold Regions Science and Technology 20, no. 2 (February 1992): 119-139.
4. Acknowledgment
The authors gratefully acknowledge the support of the Arctic Program,
Office of Naval Research, who made it possible for us to conduct these
measurements on the CEAREX program. This work was funded under contracts
N00014-89-WM-24004 and N00014-90-MP-24051.
December 1989 (rev. 8/91)
