Data Set ID: 
NSIDC-0591

WAIS Divide Ice Core Electrical Conductance Measurements, Antarctica, Version 1

This data set contains electrical measurements that were used to develop the WDC06A-7 timescale.

NSIDC does not archive these data.

Parameter(s):
  • Ice Core Records > Electrical Properties > ELECTRICAL PROPERTIES
  • Ice Core Records > Electrical Properties > ICE CORE RECORDS
  • Snow/Ice > Snow Stratigraphy > SNOW STRATIGRAPHY
Data Format(s):
  • ASCII
Spatial Coverage:
N: -79.481, 
S: -79.481, 
E: -112.1115, 
W: -112.1115
Platform(s):NIC
Spatial Resolution:Not SpecifiedSensor(s):ECM
Temporal Coverage:
  • 1 January 2007 to 31 December 2011
Version(s):V1
Temporal ResolutionNot specifiedMetadata XML:View Metadata Record
Data Contributor(s):T. Fudge, Kendrick Taylor

Geographic Coverage

Please contact the data provider for the correct Data Citation for this data set.

Literature Citation

As a condition of using these data, we request that you acknowledge the author(s) of this data set by referencing the following peer-reviewed publication.

Back to Top

Collapse All / Open All

Detailed Data Description

The West Antarctic Ice Sheet (WAIS) Divide Ice Core (WDC), in central West Antarctica, is unique in coming from a location that has experienced minimal elevation change, is strongly influenced by marine conditions, and has a relatively high snow–accumulation rate, making it possible to obtain an accurately dated record with high temporal resolution. Drilling of the WDC was completed in December 2011 to a depth of 3,405 m. Drilling was halted ~50 m above the bedrock to avoid contaminating the basal water system. WDC is situated 24 km west of the Ross–Amundsen ice-flow divide and 160 km east of the Byrd ice–core site. Refer to Figure 1. (WAIS Divide Project Members 2013)

WDC loacation map
Figure 1. Map of West Antarctica
Locations of the WDC, Byrd, and Siple Dome ice cores and the Ohio Range (OR) and Mt. Waesche (MW). Ice shelves are shown in gray; Siple Coast and Amundsen Sea ice streams are shown by blue shading. Contour interval is 500 m.

The elevation is 1766 m; the present-day snow accumulation rate is 22 cm yr–1 (ice equivalent) and the average temperature is approximately –30°C. The age of the oldest recovered ice is ~68 kyr. The WDC06A–7 timescale is based on the identification of annual layers to 29.6 kyr ago using primarily electrical measurements. To validate WDC06A–7, we compare times of abrupt changes in atmospheric methane concentration with the Greenland Ice Core Chronology 2005 (GICC05). We also compare the methane variations in WDC with abrupt changes in a speleothem δ18O record from Hulu Cave, China. The difference in age between the ice and gas at a given depth is calculated using a steady–state firn–densification model and is always less than 500 yr. The age differences between WDC06A–7 and GICC05 and between WDC06A–7 and the Hulu Cave timescale are much less than the independent timescale uncertainties. Refer to Figure 2. (WAIS Divide Project Members 2013)

age difference
Figure 2. Timescale Comparison

The WDC06A–7 timescale is based on high-resolution (>1 cm) measurements of sulphur, sodium, black carbon and Electrical Conductivity Measurements (ECM) above 577 m (2,358 yr before present (bp; ad 1950). Below 577 m WDC06A–7 is based primarily on electrical measurements: di-electrical profiling was used for the brittle ice from 577 to 1,300 m (to 6,063 yr bp). Alternating Current ( AC-ECM) were used from 1,300 to 1,955 m (to 11,589 yr bp) and both alternating-current and direct–current ECM measurements were used below 1,955 m. The interpretation was stopped at 2,800 m because the expression of annual layers becomes less consistent, suggesting that all years may not be easily recognized (WAIS Divide Project Members 2013).

The upper 577 m of the timescale was compared with volcanic horizons dated on multiple other timescales, the uncertainty at 2,358 yr bp is ± 19 yr. For the remainder of the timescale, we assigned an uncertainty based on a qualitative assessment of the clarity of the annual layers. For ice from 577 to 2,020 m (2–12 kyr ago), we estimated a two percent uncertainty based on comparisons between the ECM and chemical (Na, SO4) interpretations between 577 and 1,300 m, which agreed to within one percent. Refer to Figure 3. The estimated uncertainty increased during the deglacial transition owing to both thinner layers and a less pronounced seasonal cycle. We compared the annual-layer interpretation of the ECM records in an 800–yr overlap section (1,940–2,020-m depth, corresponding to 11.4–12.2 kyr ago) with various high-resolution chemistry records (sodium and sulphur). We found overall good agreement (19 yr more in the ECM–only interpretation) but did observe a tendency for the ECM record to split one annual peak into two small peaks. We used this knowledge in the annual–layer interpretation of the ECM record. We increased the uncertainty to four percent between 2,020 and 2,300 m (12.2–15.5 kyr ago) and to eight percent between 2,300 and 2,500 m (15.5–20 kyr ago). The glacial period had a stronger annual-layer signal than the transition, and we estimate a six percent uncertainty for the rest of the glacial. The 150–yr acid deposition event, first identified in the Byrd ice core, was found in WDC at depths of 2,421.75 to 2,427.25 m. Because there is consistently high conductance without a clear annual signal, we used the average annual layer thickness of the 10 m above and below this section to determine the number of years within it. There are periods of detectable annual variations within this depth range, and they have approximately the same annual–layer thickness as the 10-m averages. A 10 percent uncertainty was assumed (WAIS Divide Project Members 2013).

brittle ice annual signals
Figure 3. Brittle Ice Annual Signals
Comparison of electrical (dielectric profiling) and chemical (non-sea-salt sulfate) annual signals in a section of brittle ice that was rated very poor, the worst rating for ice quality. Blue triangles are annual layer picks.

We assess the accuracy of WDC06A–7 by comparing it with two high–precision timescales: GICC05 and a new speleothem timescale from Hulu Cave. Because the age of the gas at a given depth is less than that of the ice surrounding it, we first need to calculate the age offset (Δage). We use the inferred accumulation rates and surface temperatures estimated from the δ18O record constrained by the borehole temperature profile in a steady-state firn-densification model. The model is well–suited to WDC because it was developed using data from modern ice–core sites that span the full range of past WDC temperatures and accumulation rates. We calculate Δage using 200-yr smoothed histories of surface temperature and accumulation rate, a surface density of 370 kg m–3 and a close-off density of 810 kg m–3. Refer to Figure 4. The calculated present-day Δage is 210 yr, which is similar to the value, 205 yr, measured for WDC. The steady-state model is acceptable for WDC because the surface temperature and accumulation rate vary more slowly than in Greenland. Because our primary purpose is to assess the accuracy of the WDC06A-7 timescale, calculation of Δage to better than a few decades is not necessary. The Δage uncertainty between 15 and 11 kyr ago is estimated to be 100 yr. The Δage uncertainty is estimated to be 150 yr for times before 20 kyr ago because of the colder temperatures and lower and less certain accumulation rates (WAIS Divide Project Members 2013).

Assessment of WDC06A-7 Timescale via atmospheric methane
Figure 4. Assessment of WDC06A-7 Timescale via Atmospheric Methane
A) Age for WDC calculated with Herron and Langway 27 rn densication model using the inferred accumulation rate and temperature histories. Circles are the age values at the times of abrupt variations in methane used to compare timescales. 
B and C) WDC methane on WDC06A-7 minus Δage and Greenland methane composite on GICC05 gas timescale 33 . Red vertical bars correspond to the times of the red circles in A where the age was calculated . The age difference between the WDC and Greenland timescales are shown in Figure 2.

Because methane is well mixed in the atmosphere and should have identical features in both hemispheres, we use atmospheric methane measurements from WDC and the Greenland composite methane recordto compare WDC06A–7 and GICC05 at six times. The age differences are summarized in Figure 2 and the correlation and Δage uncertainties are shown in Table 1 and Table 2. In Greenland, methane and δ18O changes are nearly synchronous, and we therefore assume no Δage uncertainty in the Greenland gas timescale at times of abrupt change. An exception is at 24 kyr ago (Dansgaard–Oeschger Event 2), when methane and δ18O changes do not seem to be synchronous. We estimate the correlation uncertainty from the agreement of the methane records in Figure 4 (WAIS Divide Project Members 2013).

Table 1. Uncertainties in Timescale Comparisons WD to Greenland
Event
Approximate Age
WD Δage Uncertainty
Greenland Composite Δage Uncertainty
WD to Greenland Correlation Uncertainty
WD to Greenland Total Uncertainty
8.2 ka
8.2
50
0
50
100
YD Termination
11.7
100
0
100
200
YD Initiation
12.9
100
0
100
200
Boiling Warming
14.7
100
0
100
200
DO2
23
150
150
300
600
DO3
28
150
0
100
250
DO4
29
150
0
100
250
Table 2. Uncertainties in Timescale Comparisons WDC to Hulu
Event
Approximate Age
WDC Δage Uncertainty
WDC to Sofular Correlation Uncertainty
WDC to Sofular Total Uncertainty
WDC to Hulu Correlation Uncertainty
WDC to Hulu Total Uncertainty
Boiling Warming
14.7
100
100
200
100
200
DO3
28
150
200
350
100
250
DO4
29
150
200
350
100
250

Speleothems can be radiometrically dated with U/Th and have smaller absolute age uncertainties than do annually resolved timescales in the glacial period. Records of speleothem δ18O show many abrupt changes that have been tied to the Greenland climate record. However, the physical link between δ18O variations in the caves and methane variations is not fully understood. Therefore, there is an additional and unknown correlation uncertainty in these comparisons. We compare WDC06A–7 with the new record from Hulu Cave, China, which is the best-dated speleothem record during this time interval. Comparisons can be made at only three times; our best estimate of the age differences is 100 yr or less (WAIS Divide Project Members 2013).

Figure 5 is a visual look at the data (click on the image for a full PDF view). The images are of multi-track electrical measurement data for depths below 1956 m. Images are approximately to scale. Data for individual sections can be obtained by contacting T.J. Fudge at tjfudge@uw.edu.

Warm colors are high electrical conductivity. Cool colors are low electrical conductivity. Each track is normalized by subtracting the mean and dividing by the standard deviation. Plotted values are a 3-measurement (3mm) running average. Measurements affected by breaks in the core were masked out. X-axis is approximate horizontal position on the ice core as measured from left from looking from bottom to top of the core. Y-axis is depth in meters. Title is the tube number.

Note: Click on the image for full PDF view.

Multi-track Electrical Measurement Data Images
Figure 5. Multi-track Electrical Measurement Data Images
Format

Data are provided in tab-delimited ASCII format.

Background color on
File and Directory Structure

Data are available on the FTP site in the ftp://sidads.colorado.edu/pub/DATASETS/AGDC/nsidc0591_fudge/ directory. Within this directory, there are 108 text files. The text files that have not been normalized contain two columns of data. The first column contains depth in meters, and the second column contains conductance (S * 10^8). The text files that have been normalized contain five columns, Depth, Adjusted Data, Normalized Data, Raw Data, and Spline Fit. Text files contained 50 m of electrical measurements that have 1 mm spacing. Data at core breaks were removed.

Background color on
File Naming Convention

This section explains the file naming convention used for this product with an example.

Example File Name: 0100R.D50

xxxxR.D50

Refer to Table 3 for the valid values for the file name variables listed above.

Where:

Table 3. File Naming Convention
Variable
Description
xxxx
Starting depth.
R
Measurements that have not been normalized.
D
Dielectric Profiling Measurements (DEP).
50
File contains 50 meters of data.

This section explains the file naming convention used for this product with an example.

Example File Name: 0600S.d50

xxxxS.d50

Refer to Table 4 for the valid values for the file name variables listed above.

Where:

Table 4. File Naming Convention
Variable
Description
xxxx
Starting depth.
S
Measurements that have not been normalized, and the depths measured in the brittle ice have been shifted to match the depths measured at National Ice Core Lab (NICL); however, off sets of up to tens of cm may exist.
D
Dielectric Profiling Measurements (DEP).
50
File contains 50 meters of data.

This section explains the file naming convention used for this product with an example.

Example File Name: A1350R.A50

xxxxR.A50

Refer to Table 5 for the valid values for the file name variables listed above.

Where:

Table 5. File Naming Convention
Variable
Description
xxxx
Starting depth.
R
Measurements that have not been normalized.
A
Alternating Current - Electrical Conductivity Measurements (AC-ECM).
50
File contains 50 meters of data.

This section explains the file naming convention used for this product with an example.

Example File Name: E1350R.E50

xxxxR.E50

Refer to Table 6 for the valid values for the file name variables listed above.

Where:

Table 6. File Naming Convention
Variable
Description
xxxx
Starting depth.
R
Measurements that have not been normalized.
E
Direct Current - Electrical Conductivity Measurements (DC-ECM).
50
File contains 50 meters of data.

This section explains the file naming convention used for this product with an example.

Example File Name: a1950_adj.a50

xxxx_adj.a50

Refer to Table 7 for the valid values for the file name variables listed above.

Where:

Table 7. File Naming Convention
Variable
Description
xxxx
Starting depth.
adj
Measurements that have been normalized and processed with the procedures outlined in the Data Acquisition and Processing section of this document.
a
Alternating Current - Electrical Conductivity Measurements (AC-ECM).
50
File contains 50 meters of data.

This section explains the file naming convention used for this product with an example.

Example File Name: e1950_adj.e50

xxxx_adj.e50

Refer to Table 8 for the valid values for the file name variables listed above.

Where:

Table 8. File Naming Convention
Variable
Description
xxxx
Starting depth.
adj
Measurements that have been normalized and processed with the procedures outlined in the Data Acquisition and Processing section of this document.
e
Direct Current - Electrical Conductivity Measurements (DC-ECM).
50
File contains 50 meters of data.
Background color on
File Size

Files range from 357 KB to 2965 KB.

Background color on
Spatial Coverage

WAIS Divide, Antarctica: 79.4810° S, 112.1115° W

Spatial Resolution

Electrical measurements to 3400 m depth of the WAIS Divide Ice Core

Background color on
Temporal Coverage

Data were collected from January 2007 to December 2011.

Background color on
Parameter or Variable

Depth (m)
Conductance (uS)

Parameter Description

Direct Current - Electrical Conductivity Measurements (DC-ECM): 1000 VDC, electrode 1 cm apart, 3 mm diameter gold

Alternating Current - Electrical Conductivity Measurements (AC-ECM): 100KHZ, 2v, electrode 1 cm apart, 5x5 mm, brass

Dielectric Profiling Measurements (DEP): 100KHz, 2v, curved electrodes fitted around netted core

Sample Data Record

Figure 6 is sample data from the 0100R.D50 data file.

sample data record
Figure 6. Sample Data Record
Background color on

Software and Tools

Background color on

Data Acquisition and Processing

Multi-track DC-ECM Processing

The DC-ECM data is an average of multiple tracks along the same core but at different horizontal positions across the core. The seven tracks were averaged after the high and low value at each measurement position was excluded. For 2750-2800 meters, there was concern that the greater layer dips would adversely influence the stacking. To correct for the lay dips, the tracks were first aligned by calculating the layer inclination from the correlation of the two outside tracks and shifting each track up or down by the appropriate distance. Then the tracks were stacked in the same way as for 1955-2750 meters.

Alignment of AC and DC Data

The DC and AC data were often not properly aligned because the inclination of the layers resulted in mm scale depth offsets. To align the DC-ECM and AC-ECM, the AC depths were shifted to the position of maximum correlation with the DC data.

Data Adjustments

Different normalization techniques were used for different segments of the data: The DEP data for depths 570-1350 m are normalized. The normalized data was obtained by subtracting the mean value of each meter and then dividing by the average of the absolute value of the difference. This corrected for meter-to meter differences in ice temperature when it was measured and electrode contact issues. This method has the disadvantage of reducing the amplitude of annual layers near volcanic peaks.

The AC-ECM data for depths 1350-1950 meters were not adjusted.

The AC- and DC-ECM for depths 1950-2800 meters were normalized and adjusted. Unnormalized and multi-track data will be made available soon.

Normalization: The data were normalized by subtracted the mean conductivity of the section, then dividing by three times the standard deviation. The mean and standard deviations were calculated from a subset of the data to prevent volcanic signals from obscuring the annual signal after the data was normalized. The data 9 cm to either side of the peak value and the lowest five percent were excluded from the mean and standard deviation calculations. This exclusion was done to reduce the effect of volcanic peaks or dust lows on the amplitude of the annual signal.

Adjustment: The normalized data were then fitted with a cubic smoothing spline (Matlab script CFSMTHSPL.m which is based on CSAPS.m). The purpose of the smoothing spline was to correct lower values that commonly occurred at the ends of cores because of the position of the core tray on the ECM tracks. Only splines with a stiff spline parameter (typically 0.5) were applied. The spline curved never had more than two inflections points so the wavelengths of the spline were much longer the annual layer signal in the data.

Background color on

References and Related Publications

Contacts and Acknowledgments

Tyler J. Fudge
University of Washington
Department of Earth and Space Sciences
Johnson Hall Room 070
Box 351310
4000 15th Avenue NE
Seattle, WA 98195-1310

Kendrick Taylor
Desert Research Institute
Department of Earth and Space Sciences
Division of Hydrological Sciences
Nevada System of Higher Education
Reno, Nevada 89512

Acknowledgments: 

This research was supported by NSF OPP Grant Numbers 0944197, 0944348, 0944191, 0440817, 0440819, and 0230396.

Document Information

Document Creation Date

August 2014

No technical references available for this data set.
No FAQs or How Tos available for this data set.

Access complete Knowledge Base

Questions? Please contact:
NSIDC User Services
Phone: 1 303 492-6199
Email: nsidc@nsidc.org