Capacitively Coupled Resistivity Survey of the Sea Ice Near Barrow, Alaska

1
Capacitively Coupled Resistivity Survey of the
Sea Ice Near Barrow, Alaska Dr. Rhett Herman,
James Inman Department of Chemistry and Physics ,
Radford University, Virginia 24142
Northern end of survey line
Figure 1. Electrical resistivity cross-section of
the sea ice near Barrow, Alaska, March 2006, with
snow depth data superimposed. Note that snow
depths were taken approximately 2/3 meter to the
(seaward) side of the resistivity data. Snow
depth data courtesy of Dr. Julienne Stroeve
(National Snow and Ice Data Center,
http//nsidc.org/ ). All distances are in meters.
BACKGROUND
ABSTRACT
RESULTS AND DISCUSSION
Capacitively coupled resistivity methods have the
ability to image areas of high resistivity such
as the arctic sea ice. Due to their mobility,
capacitive arrays typically take data faster than
other resistivity methods and can thus cover a
wider survey area in a given amount of time.
This poster shows the preliminary results of a
new survey carried out near Barrow, Alaska using
the OhmMapper capacitively coupled resistivity
survey. The data was acquired along a 300-meter
line on the Chukchi Sea ice just offshore from
the BASC research station.1 This
proof-of-concept survey demonstrates the ability
of the capacitively coupled array to provide a
resistivity image of the sea ice showing features
including the uppermost snow layer and the
ice/water boundary.
Past surveys of arctic ice thickness have used
the EM-31 induction system. Conductivity readings
were obtained along with borehole measurements of
ice thickness to create calibration curves
relating the signal and the ice thickness.2
This method is effective for determining average
overall sea ice thickness but can not determine
thickness of snow layer. Snow depths have been
determined by hand using at first calibrated
poles and more recently using Magnaprobes
3. Surveys using capacitive systems in the
arctic have been carried out across permafrost
terrain (e.g., Ref. 4) using dipole-dipole
spacings on the order of tens of meters, giving
depths of penetration of the same magnitude and
showing features on the order of meters in
extent.

• We used RES2DINV to invert our data and produce
  the resistivity cross-section between the depths
  z0.21m-2.72m in Fig. 1. 6 A number of features
  may be seen, with just a few highlighted below.
• The low-resistivity seawater (blue) transitions
  into the medium-resistivity ice (yellow/orange)
  and the high-resistivity snow cover
  (orange/purple). The undulations of the ice/water
  boundary are seen all along the survey line. The
  ice is clearly thicker on the southern (left) end
  of the line, and thinner on the northern end, and
  matches a private communication of the results of
  the CRREL- and NSIDC-led team from their previous
  weeks survey of the same line.
• Magnaprobe data 3 of the snow depths were taken
  along the same line, but approximately 2/3 meter
  to the seaward side of our survey line (we did
  not want to walk directly over the test lines
  survey stakes). This data is plotted on the same
  horizontal and vertical scale as our data, and is
  seen superimposed on our cross-section. The
  Magnaprobe data follows the resistivity contours.
  
• There is an area at x241m where we often had
  some difficulty with very low-resistivity
  readings, forcing us to walk even more slowly
  across this segment. Upon inspection of the image
  above, this seems to be a large separation in the
  ice reaching towards the surface, allowing the
  seawater upward.
• There is a feature at x148m that is consistent
  with a fracture in the ice. This is consistent
  with the areas ice having been recently reformed
  after being broken up in January 2006 by ice
  forced up through the Bering Strait by wind/ocean
  currents.

ONGOING AND FUTURE WORK
Figure 2. Same resistivity data as Fig. 1 but
with a sharper color distinction between the
various resistivities to sharpen the ice/water
boundary.
We are still processing the data obtained in
March, 2006. For example, the image in Figure 2
used the same resistivity data with a sharper
distinction between the various resistivities to
highlight the ice/water boundary. This image
could be used to calculate the true structure and
volume of the sea ice at these small scales.
Data was also taken in a 2-dimensional grid,
size100m by 10m, with n0.25, 0.50, 0.75, 1.00,
1.25. Taking data in this 2-d grid at multiple
depths will allow for 3-dimensional data
inversion. This will yield a 3-d volumetric image
of the sea ice at the same sub-meter scales as
the cross-sections above. More trips to the sea
ice for longer and for more 3-D surveys are
planned for the coming year.
METHOD
This 300-meter-long survey was performed using
the OhmMapper capacitive array in its typical
dipole-dipole spacing. The data is acquired
automatically at 1.0 sec intervals the linear
data density is determined by the speed of the
operator towing the array along the surface. Our
survey speed was 1/3 m/s, giving 3 data points
per horizontal meter. The effective penetration
depth of the 16.5 kHz signal is determined by the
separation between the transmitter (Tx) and
receiver (Rx) dipoles (see side panel for
illustration). Our dipoles were L5.0m long. The
term n-spacing refers to the length of the rope
separating the ends of the Tx and Rx . A spacing
of n1.0 means the rope is 1.0 times the length
of the Tx (or Rx) dipole.
Typical values for n-spacings in other are n1.0,
2.0, 3.0, etc. This survey used very small
n-spacings in order to keep the signal as shallow
as possible and image the one- to two-meter thick
sea ice. N-spacings used in this survey were
n0.25, 0.50, 0.75, 1.00 and 1.25. Our set of
small n-spacings extensively probed the shallow
ice. For example, our L5.0 meter dipoles, the
effective penetration depth e.g. for n1.0 would
be z0.42L2.1 meters Loke pdf. The 300-m
line was surveyed twice (different days), once
with Radford Universitys Tx, and once with
another, newer Tx sent to Barrow by Geometrics.
The processed images from these two data sets
were indistinguishable.
References

• The authors would like to thank the members of
  the team consisting of members of CRREL, NSIDC,
  et al team for permission to use the test line
  they surveyed immediately prior to our survey.
• Kovacs, A., Diemand, D., and Bayer, J. Jr., 1996,
  Electromagnetic induction sounding of sea ice
  thickness. CRREL Report 96-6.
• The authors would like to thank Dr. Juliene
  Stroeve of the Dr. Julienne Stroeve of the
  National Snow and Ice Data Center
  (http//nsidc.org/ ) for permission to use her
  Magnaprobe data on the Chukchi test line.
• Calvert, H. T., Capacitive-coupled resistivity
  survey of ice-bearing sediments, Mackenzie Delta,
  Canada. Geological Survey of Canada, 2002.
• Loke, M. L., Electrical imaging surveys for
  environmental and engineering studiesa practical
  guide to 2-D and 3-D surveys, Minden Heights,
  Malaysia, 1999. Distributed by http//www.geoelect
  rical.com.
• The authors would like to thank Geometrics, Inc.
  for assistance in data analysis.
• Kovacs, A., Valleau, N., and Holladay, J. S.,
  1987, Airborne electromagnetic sounding of sea
  ice thickness and sub-ice bathymetry. CRREL
  Report 87-23.