On Wednesday, September 27, from 9:30 a.m. to 12:00 p.m./noon (USA Mountain Time), AMSR-E, Aquarius, IceBridge, ICESat/GLAS, MEaSUREs, MODIS, NISE, SMAP, and VIIRS data will not be available due to system maintenance.

Get Data

DOWNLOADING DATA VIA FTP

Data can be downloaded through a Web browser or command line via FTP. When using a Web browser, the FTP link first directs you to an Optional Registration Form that if filled out, will allow you to receive notifications about updates or processing changes related to that specific data set. After completing the Optional Registration Form, the FTP directory becomes available. For additional help downloading data through an FTP client, go to the How to access data using an FTP client support page.

FTP

Data Set ID:
G02177

Ice Draft and Ice Velocity Data in the Beaufort Sea, 1990-2003, Version 1

This data set provides measurement of sea ice draft (m) and the movement of sea ice (cm/s) over the continental shelf of the Eastern Beaufort Sea. The data set spans April 1990 to September 2003. The data set was acquired to assist in determining the character and recurrence of ice features hazardous to offshore industrial activity, specifically for the exploitation of hydrocarbon reserves and the impact of climate variability and change on the Beaufort Sea ice regime. Observations were made using two types of self-contained Upward Looking Sonar (ULS) moored near the sea floor: an Ice Profiling Sonar (IPS) used to obtain ice draft data and an Acoustic Doppler Current Profiler (ADCP) used to obtain the ice velocity data.

Geographic Coverage

Parameter(s):
  • Sea Ice > Ice Draft
  • Snow/Ice > Ice Velocity
Spatial Coverage:
  • N: 74.152, S: 70.083, E: -125.577, W: -133.828

Spatial Resolution:
  • Varies
Temporal Coverage:
  • 1 April 1990 to 30 September 2003
Temporal Resolution: 4 minute, 30 minute
Data Format(s):
  • ASCII Text
Platform(s) BUOYS
Sensor(s): ADCP, UPWARD LOOKING SONAR
Version: V1
Data Contributor(s): Humfrey Melling, D Riedel
Data Citation

As a condition of using these data, you must cite the use of this data set using the following citation. For more information, see our Use and Copyright Web page.

Melling, H. and D. A. Riedel. 2008. Ice Draft and Ice Velocity Data in the Beaufort Sea, 1990-2003, Version 1. [Indicate subset used]. Boulder, Colorado USA. NSIDC: National Snow and Ice Data Center. doi: http://dx.doi.org/10.7265/N58913S6. [Date Accessed].

Back to Top

Collapse All / Open All

Background Information

Note: The majority of this document was created from Draft and Movement of Pack Ice in the Beaufort Sea: A Time-Series Presentation April 1990 - August 1999 (H. Melling and D.A. Riedel 2004). Download this document without graphics: draft_movement_beaufort_2004.pdf (PDF, 1.6 MB). Download this document in zip format with graphics: draft_movement_beaufort_2004.zip (zip file, 4.7 MB).

Sea ice grows in vast featureless sheets of uniform thickness if not disturbed by wind and current. An equilibrium thickness is attained when the heat conducted up through the ice from the ocean is adequate to balance the energy loss at the atmospheric interface. As this budget varies seasonally, so does the equilibrium thickness. Calculations show that the annually averaged thickness of Arctic sea ice, when subject only to thermodynamic forcing, is about 3 meters (Maykut and Untersteiner 1971).

The constant movement of sea ice in response to winds and currents generates stresses which cause ice sheets to break apart into floes separated by leads of open water. Under cold conditions, these leads develop a new cover of thinner ice. Where moving floes collide, ice is broken into fragments and piled into sinuous mounds called ridges. By these processes, the ice pack is quickly deformed into a rough and geometrically complex landscape both above and below the sea surface. In the southern Beaufort Sea in late winter, first-year sea ice typically ranges in thickness from zero to about two meters. However, ridges accumulate to a much greater total thickness. The deepest free-floating ridge keel recorded to date almost extended to a depth of 50 meters (Kovacs et al. 1973).

This study is focused in the Eastern Beaufort Sea. Its initial motivation was the acquisition of a statistical description of pack ice in the area, particularly in relation to features of extreme draft (ridge keels) and their rate of drift. From 1976 to 1987 there was very active exploration for oil within the zone of drifting pack ice. Since 2001 there has been a resurgence of interest in exploration, with emphasis this time on natural gas within the zone of land-fast ice.

Pressure ridges are the most severe commonplace ice hazard to offshore structures and shipping. Icebreaking ships and drilling platforms typically reach their design limits when transiting ice ridges. Grounded ridges gouge deeply into the seabed in the Beaufort Sea, thereby threatening sub-sea well completions and pipelines. Accurate data are needed for the cost-effective and safe design of the offshore infrastructure such as surface piercing platforms, seabed installations, sub-sea pipelines, icebreaking ships, environmental constraints on operations, etc.

Observations were made using two types of self-contained sonar moored near the seafloor. A four-beam Doppler sonar measures the velocity of ice drift and a narrow-beam ice-profiling sonar measures its draft. Ten sites on the Mackenzie and Banks Island shelves were instrumented to meet various objectives during this period. However, long time series were maintained at only three locations, namely the middle shelf and shelf edge north of the Mackenzie delta and the shelf edge to the north-west of Amundsen Gulf. In a typical year, pack ice covers the sites except in late summer. Approximately 2000 km of pack ice were surveyed annually by each installation.

Detailed Data Description

Parameters

The parameters of this data set include sea ice draft (m) and sea ice velocity/ice drift (cm/s). Ice draft is a measurement of the thickness of the sea ice below the waterline and often serves as a close proxy for total ice thickness. The ice velocity or ice drift is a measure of the speed of the drift of the ice.

Background color on
Spatial and Temporal Coverage and Resolution

This data set covers the Eastern Beaufort Sea. Figure 1 is a graphical representation of the mooring sites; see Table 11 for a complete list of specific latitudes and longitudes by mooring site number.

Mooring Sites
Figure 1. Sonar Mooring Sites (Click for high resolution image)

Temporal Coverage and Resolution

This data set begins in April 1990 and spans through September 2003. Nominal temporal resolutions are four minutes for ice draft data, 30 minutes for ice drift data. Note: Actual values vary somewhat from deployment to deployment. Statistics were calculated for nominal monthly intervals.

Background color on
Format

All files are provided in ASCII text format. The top line of all data and header files is a time stamp of the date the file was created. The data files (both ice draft and ice velocity) contain flags as described in Table 1.

Table 1. Data Flag Descriptions
Value Description
0 Inserted when the original full length sonar record revealed prolonged ice free conditions or rough seas in which ice at low concentration cannot be detected. Note: Shorter periods of open water were not removed during processing and are evident as ice draft near zero (±0.05 m)
-88.88 Inserted when no observations were made. For example, there was no instrument in the sea or the instrument was non-functional.
-99.99, -99 Inserted when temporary ambiguity in the validity of echo ranges merited a judgment of bad data.

Header Files

Each data file for both the ice draft and ice velocity data is paired with an ice draft header file in ASCII text format that provides specific information about its data file. Each header file contains information about the location of the moored sonar and other log information. The headers have a standardized set of sections, sub-sections, and fields within sub-sections. However, these components can be in any sequence and not all are included in every header file. In the files, the * symbol indicates the start of a section and the $ symbol delineates the start and end of a sub-section.

Ice Draft Files

Data Files

The ice draft data files are provided in ASCII text format and contain four columns of data as described in Table 2.

Table 2. Column Descriptions for Ice Draft Data Files
Column Description
1 Decimal day of the measurement relative to 00:00 UTC on January 1, 1990 (which has a value of 1.000)
Note: The header files incorrectly list this column as being Julian days.
2 Decimal day of the measurement relative to 00:00 UTC on January 1 of the year of instrument deployment
Note: The header files incorrectly list this column as being Julian days.
3 Ice Draft (m)
4 4-digit year that the instrument was deployed

Statistics Files

The statistics files contain various statistical indicators of ice draft and are calculated over monthly intervals. The monthly intervals alternate between 31-day and 30-day durations based on the calendar year. In a leap year, the last "month" of the year has 30 days; otherwise it has 29 days.

Table 3 describes the rows in the ice draft statistics files.

Table 3. Row Descriptions for Ice Draft Statistics Files
Row Description
Section 3-character month and 2-digit year of the section being analyzed
Start rec Start number of the record
End rec End number of the record
Total rec Total number of records available for statistical analysis
Flags Number of flags present in the record
Total possible rec Total number of possible records
Mean (from D_S) This is the mean computed from all non-flagged data values including open water (zero-draft values are included).
StDev (from D_S) This is the standard deviation computed from all non-flagged data values including open water (zero-draft values are included).
Mean from HIS Mean from the histogram: The mean calculated as the sum of the products of the bin-center-value times the bin population. This calculation is included to allow direct comparison with the mean with out open water.
Mean w/o O/W Mean excluding open water. The mean calculated as the sum of the products of the bin-center-value times the bin population excluding the first bin.
Ice Concentration Percent ice concentration
Fraction of possible Fraction of the total possible records available for statistical analysis
Min Minimum value
20th %ile 20th percentile
50th %ile 50th percentile (median)
80th %ile 80th percentile
Max Maximum value
BIN(m) Binned draft data. There are 401 bins with the smallest spanning (-0.05, 0.05) and the largest (39.95, 40.05) m.

Ice Velocity Files

Data Files

The ice velocity data files are provided in ASCII text format and contain three columns of data as described in Table 4.

Table 4. Column Descriptions for Ice Velocity Data Files
Column Description
1 Decimal days relative to the start time of the instrument 
Note: The header files incorrectly list this column as being Julian days.
2 Ice speed (cm/s)
3 Direction towards which ice is drifting in degrees in a geographic convention referenced to true North and increasing clockwise from North.

Statistics Files

The statistics files contain various statistical indicators of ice velocity for nominal monthly intervals. The intervals alternate between 31-day and 30-day durations based on the calendar year. In a leap year, the last "month" of the year has 30 days; otherwise it has 29 days.

Table 5 describes the rows in the ice speed statistics files.

Table 5. Row Descriptions for Ice Speed Statistics Files
Row Description
Section 3-character month and 2-digit year of the section being analyzed
Start rec Start number of the record
End rec End number of the record
Total rec Total number of records available for statistical analysis
Flags Number of flags present in the record
Time increment The velocity sampling interval in minutes. Each value is a vector average over the interval. Times are referenced to the start of the averaging interval.
Total possible rec Total number of possible records
Mean (from D_S) The mean computed from all non-flagged data values.
StDev (from D_S) The standard deviation computed from all non-flagged data values.
Mean from HIS Mean from the histogram: The mean calculated as the sum of the products of the bin-center-value times the bin population. This calculation is included to allow direct comparison with the mean when moving.
Mean when Moving The mean of all non-zero ice speeds. This variable is included because during winter ice can can be stationary approximately 50 percent of the time.
Percentage no-motion Percentage of the record with ice speed of 0 cm/s
Percentage of possible Percentage of the total possible records available for statistical analysis
Min Minimum value
20th %ile 20th percentile
50th %ile 50th percentile (median)
80th %ile 80th percentile
Max Maximum value
BIN(cm/s) Binned ice speed data. The bins of the histogram are 2 cm/s in width. There are 50 bins with the smallest spanning (0, 2) cm/s and the largest (98, 100) cm/s.
Background color on
File and Directory Structure

The data are available via FTP and are divided into two directories: Statistics and TimeSeries. These two directories are further subdivided and are described in Table 6 and Figure 2.

Table 6. File and Directory Structure
Directory Description
Statistics Contains the statistics files for this data set. This directory is further subdivided into two other directories: IceDraft and IceSpeed. These two directories contain the ice draft and ice speed statistics files, respectively.
TimeSeries Contains the data and header files for this data set. This directory is further subdivided into two other directories: IceDraft and IceVelocity. These two directories contain the ice draft and ice velocity data files, respectively.

Figure 2 displays the FTP directory structure.

Directory Structure
Figure 2. FTP Directory Structure
Background color on
Sample Data Record

Ice Draft Data Files

The following is a sample data record of the ice draft data files. This sample shows the first six lines of Beaufort_site01_1992-04_draft.data. See Table 2 for a description of the columns.

 846.87500 116.87480   1.62 1992
 846.87781 116.87760   1.62 1992
 846.88055 116.88040   1.61 1992
 846.88330 116.88320   1.59 1992
 846.88611 116.88600   1.61 1992
 846.88892 116.88870   1.61 1992

Ice Velocity Data Files

The following is a sample data record of the ice velocity data files. This sample shows the first six lines of Beaufort_1990-03_drift.data from Site 01. See Table 4 for a description of the columns.

  90.37150      0.2    314.1
  90.39342      0.5    287.3
  90.41534      0.8    280.1
  90.43725      1.7    347.3
  90.45917      1.7    305.3
  90.48109      3.1    308.9
Background color on
File Naming Convention

Ice Draft Data and Header Files

The ice draft data and header files are named according to the following convention and as described in Table 7:

Beaufort_siteXX_YYYY-MM_draft.FFFFF

Where:

Table 7. File Naming Convention for Ice Draft Data and Header Files
Variable Description
Beaufort Identifies this as data from the Beaufort Sea
XX 2-digit site number (01, 02, 03, 04, 05, 06, 07, 08, or 09) 
See Figure 1 for a map of the sites and Table 11 for specific latitude and longitude of the sites
YYYY 4-digit year that the record starts at the site
MM 2-digit starting month in the starting year
draft Identifies this as containing ice draft data
FFFFF File type (data: data file, header: header file)

Ice Velocity Data and Header Files

The ice velocity data and header files are named according to the following convention and as described in Table 8:

Beaufort_siteXX_YYYY-MM_drift.FFFFF

Where:

Table 8. File Naming Convention for Ice Velocity Data and Header Files
Variable Description
Beaufort Identifies this as data from the Beaufort Sea
XX 2-digit site number (01, 02, 03, 04, 05, 06, 07, 08, or 09) 
See Figure 1 for a map of the sites and Table 11 for specific latitude and longitude of the sites
YYYY 4-digit year that the record starts at the site
MM 2-digit starting month in the starting year
drift Identifies this as containing ice velocity/ice drift data
FFFFF File type (data: data file, header: header file)

Statistics Files

The statistics files are named according to the following convention and as described in Table 9:

Beaufort_SiteXX_PTYPE_Stats.csv

Where:

Table 9. File Naming Convention for Statistics Files
Variable Description
Beaufort Identifies this as data from the Beaufort Sea
XX 2-digit site number (01, 02, 03, 04, 05, 06, 07, 08, or 09)
PTYPE Data parameter (drift or draft)
.csv Identifies this as a comma delimited ASCII text file
Background color on
File Size

Table 10 lists the size range of all files.

Table 10. File Size Range
File Size Range
Ice draft data files 2.3 MB to 50 MB per file
Ice draft header files 3.0 KB to 8.1 KB per file
Ice velocity data files 69 KB to 1.6 MB per file
Ice velocity header files 9.1 KB - 16 KB per file
Statistics files (ice draft and ice velocity) 4.1 KB to 239 KB per file
Background color on

Software and Tools

Software and Tools

Users are advised to use any software that can read ASCII text.

Background color on

Data Acquisition and Processing

Sensor or Instrument Description

The Upward Looking Sonars (ULS) used to acquire this data were moored to the ocean floor. At each site, an Ice Profiling Sonar (IPS) was used to obtain ice draft data; and an Acoustic Doppler Current Profiler (ADCP) was used to obtain the ice velocity data. Exceptions are the use of a Water Structure Profiler (WASP) sonar instead of IPS for ice draft observation at Site 2 in 1991 and 1992, and the deployment of IPS without ADCP at Site 9 during 2001, where the primary emphasis was on the measurement of wind waves.

The newest model of the ice profiling sonar used for data acquisition was model 4 (IPS4), developed in 1995 and first used in 1996. Model 1 (IPS1) never delivered useful data. Models 2 (IPS2) and 3 (IPS3) differed only in terms of data storage and fail-safe features in firmware. IPS4 is a version completely re-engineered to reduce energy demand, sensitivity to other targets, and cost; and to increase reliability, operating flexibility, operating endurance, and data storage. The pendulums used to measure pitch and roll by IPS2 and IPS3 were replaced with solid-state sensors in IPS4; the operating frequency was increased from 200 kHz to 420 kHz and the data storage from 16 megabytes to 66 megabytes and ultimately to 132 megabytes. IPS4 can operate sequentially in up to eight unique configurations, activated on preset dates. This feature permits optimal use of the battery and data capacity as ice conditions change throughout the year. IPS4 also has a burst mode suited for brief high-resolution surveys or wave measurement. The WASP sonar, used at Site 2, is a self-contained echo sounder developed at Institute of Ocean Sciences (IOS) to record the amplitude of back-scatter from the water column.

Table 11 lists the mooring site history and Tables 12, 13, and 14 describe the instrument type, mooring type, and location type, respectively, used in Table 11. The gray type denotes failed deployments and underlining denotes a qualified success. Entries that are neither gray nor underlined indicate recovery of good data from that deployment.

Table 11. Mooring Site History
Site Year Deployment
Name
Location Latitude
(deg min N)
Longitude
(deg min W)
Instrument Sonar (kHz) Sample Interval
(mm ss)
Mooring
type
Sonar
depth
(m)
Water
depth
(m)
1 1990 A179 ISC90-1A 70 18.300 133 36.620 NB-ADCP 614 31 34 2 50 56
1 1990 90IPS1-1 ISC90-1A 70 20.660 133 44.160 IPS1 200 15 0 54 56
1 1990 B179 ISC90-1B 70 20.830 133 43.960 NB-ADCP 614 30 2 50 56
1 1992 92IPS3-2A ISC92-1 70 19.892 133 49.664 IPS3 200 10 1 49 54
1 1992 92IPS3-2B ISC92-1 70 19.938 133 49.433 IPS3 200 10 1 48 53
1 1992 D318 ISC92-1 70 19.819 133 49.666 NB-ADCP 307 40 2 49 54
1 1993 93IPS3-2 ISC93-1 70 17.535 133 36.643 IPS3 200 10 1 50 55
1 1993 E318 ISC93-1 70 17.532 133 36.755 NB-ADCP 307 40 2 50 55
1 1994 94IPS3-1 ISC94-1 70 17.342 133 39.336 IPS3 200 10 1 49 54
1 1994 F464 ISC94-1 70 17.377 133 39.396 NB-ADCP 307 30 2 49 54
1 1995 95IPS3-1 ISC95-1 70 25.718 133 47.728 IPS3 200 10 1 55 60
1 1995 G506 ISC95-1 70 25.731 133 47.681 NB-ADCP 307 30 2 55 60
1 1996 96IPS3-1 ITT96-1 70 20.158 133 42.328 IPS3 200 6 1 52 57
1 1996 H318 ITT96-1 70 20.172 133 42.344 NB-ADCP 307 15 2 52 57
1 1997 97IPS3-2 ITT97-1 70 19.968 133 41.474 IPS3 200 5 1 49 54
1 1997 I318 ITT97-1 70 20.025 133 41.819 NB-ADCP 307 15 2 48 53
1 1998 98IPS4-0 ITT98-1 70 20.093 133 45.260 IPS4 420 2,5 1 50 55
1 1998 J586 ITT98-1 70 20.051 133 45.078 NB-ADCP 307 30 2 50 55
1 1999 99IPS3-3 ITT99-1 70 20.080 133 44.830 IPS3 200 4 1 50 55
1 1999 K318 ITT99-1 70 20.067 133 44.896 NB-ADCP 307 15 2 51 56
1 2000 00IPS3-1 ITC00-1 70 20.076 133 44.841 IPS3 200 4 1 50 55
1 2000 L464 ITC00-1 70 20.075 133 44.908 NB-ADCP 307 15 2 50 55
1 2001 01IPS4-0 ITC01-1 70 19.977 133 44.471 IPS4 420 3,1 1 50 55
1 2001 M464 ITC01-1 70 19.896 133 44.232 NB-ADCP 307 30 2 50 55
1 2002 02IPS4-0 ITC02-1 70 19.977 133 44.470 IPS3 200 3,1 1 50 55
1 2002 N464 ITC02-1 70 19.972 133 44.410 NB-ADCP 307 30 2 50 55
1 2003 03IPS4-1 ITC03-1 70 19.937 133 44.274 IPS4 420 3,1 1 52 57
1 2003 O506 ITC03-1 70 19.973 133 44.463 NB-ADCP 307 30 2 51 56
2 1990 90IPS2-1A ISC90-2A 70 48.620 133 43.410 IPS2 200 15 1 74 79
2 1990 A318 ISC90-2A 70 48.660 133 43.250 NB-ADCP 307 30 25 2 70 76
2 1990 90IPS2-1 ISC90-2B 70 58.090 133 41.500 IPS2 200 15 1 77 82
2 1990 B318 ISC90-2B 70 58.120 133 41.090 NB-ADCP 307 30 2 76 82
2 1991 91IPS1-1 ISC91-2 70 52.779 133 44.976 IPS1 200 10 0 78 80
2 1991 91WASP-1 ISC91-2 70 53.269 133 43.842 WASP 200 1 30 1 75 80
2 1991 C318 ISC91-2 70 53.178 133 43.939 NB-ADCP 307 45 2 75 81
2 1992 92IPS3-1A ISC92-2 70 57.211 133 44.076 IPS3 200 10 1 77 82
2 1992 92IPS3-1B ISC92-2 70 57.267 133 43.848 IPS3 200 10 1 78 83
2 1992 D506 ISC92-2 70 57.134 133 43.812 NB-ADCP 307 30 2 77 82
2 1993 93IPS3-1 ISC93-2 70 56.849 133 42.823 IPS3 200 10 1 73 78
2 1993 E506 ISC93-2 70 56.774 133 42.677 NB-ADCP 307 30 2 73 78
2 1994 94IPS3-2 ISC94-2 70 57.628 133 42.697 IPS3 200 10 1 74 79
2 1994 F506 ISC94-2 70 57.681 133 42.629 NB-ADCP 307 30 2 74 79
2 1995 95IPS3-2 ISC95-2 70 56.088 133 43.873 IPS3 200 10 1 72 77
2 1995 G318 ISC95-2 70 56.052 133 44.017 NB-ADCP 307 30 2 72 77
2 1996 96IPS3-2 ITT96-2 70 56.536 133 41.695 IPS3 200 6 1 74 79
2 1996 H464 ITT96-2 70 56.519 133 41.578 NB-ADCP 307 30 2 74 79
2 1997 97IPS4-0 ITT97-2 70 56.563 133 41.567 IPS4 420 3 1 74 79
2 1997 I464 ITT97-2 70 56.615 133 41.679 NB-ADCP 307 20 2 74 79
2 1998 98IPS3-2 ITT98-2 70 56.369 133 41.134 IPS3 200 5 1 75 80
2 1998 J464 ITT98-2 70 56.381 133 41.116 NB-ADCP 307 30 2 75 80
2 1999 99IPS4-1 ITT99-2 70 56.370 133 41.090 IPS4 420 2,4,3,6 1 75 80
2 1999 K506 ITT99-2 70 56.380 133 41.070 NB-ADCP 307 40 2 75 80
2 2001 01IPS4-1 ITC01-2 70 59.310 133 45.042 IPS4 420 10,6 3 48 116
2 2001 M506 ITC01-2 70 59.310 133 45.042 NB-ADCP 307 40 3 111 116
2 2003 03IPS4-0 ITC03-2 71 02.251 133 24.180 IPS4 420 4,5 3 45 113
2 2003 O586 ITC03-2 71 02.251 133 24.180 NB-ADCP 307 40 3 108 113
3 1991 91IPS3-1 ISC91-3 70 32.076 131 30.368 IPS3 200 10 1 47 52
3 1991 C506 ISC91-3 70 32.138 131 30.167 NB-ADCP 307 30 2 46 52
4 1991 91IPS3-1 71 14.276 130 59.686 IPS2 200 10 1 72 77
4 1991 C464 ISC91-4 71 14.392 130 59.629 NB-ADCP 307 30 2 71 77
4 1992 92IPS3-3A ISC92-4 71 26.986 130 13.298 IPS3 200 10 1 82 87
4 1992 92IPS3-3B ISC92-4 71 26.847 130 13.115 IPS3 200 10 1 79 84
4 1992 D464 ISC92-4 71 26.991 130 12.937 NB-ADCP 307 30 2 82 87
4 1993 93IPS3-3 ISC93-4 71 27.888 130 15.779 IPS3 200 10 1 75 80
4 1993 E464 ISC93-4 71 27.845 130 16.065 NB-ADCP 307 30 2 75 80
4 1994 94IPS3-3 ISC94-4 71 26.811 130 16.828 IPS3 200 10 1 75 80
4 1994 F586 ISC94-4 71 26.728 130 16.833 NB-ADCP 307 30 2 75 80
4 1995 95IPS3-3 ISC95-4 71 24.667 130 14.921 IPS3 200 10 1 56 61
4 1995 G586 ISC95-4 71 24.738 130 14.867 NB-ADCP 307 30 2 78 83
5 1996 96IPS3-3 ITT96-5 73 27.133 126 36.051 IPS3 200 10 3 45 108
5 1996 H586 ITT96-5 73 27.133 126 36.051 NB-ADCP 307 30 3 103 108
6 1997 97IPS4-1 ITT97-6 74 09.096 125 54.373 IPS4 420 5 3 48 85
6 1997 I506 ITT97-6 74 09.096 125 54.373 NB-ADCP 307 45 3 80 85
7 1998 98IPS3-1 ITT98-7 73 29.561 125 34.635 IPS3 200 10 1 49 54
7 1998 J318 ITT98-7 73 29.560 125 34.826 NB-ADCP 307 30 2 49 54
8 1998 98IPS4-1 ITT98-8 72 34.955 127 19.274 IPS4 420 10,5 3 45 105
8 1998 J506 ITT98-8 72 34.955 127 19.274 NB-ADCP 307 30 3 100 105
8 1999 99IPS4-0 ITT99-8 72 34.689 127 18.403 IPS4 420 8,3,6 3
8 1999 K586 ITT99-8 72 34.689 127 18.403 NB-ADCP 307 40 3 96 101
8 2001 01IPS3-2 ITC01-8 72 34.745 127 26.246 IPS3 200 5 3 50 110
8 2001 M318 ITC01-8 72 34.745 127 26.246 NB-ADCP 307 30 3 105 110
8 2003 03IPS4-33 ITC03-8 72 34.680 127 27.174 IPS4 420 2,4 3 48 113
8 2003 O318 ITC03-8 72 34.680 127 27.174 NB-ADCP 307 40 3 108 113
9 2001 01IPS4-2 BMH01-9 70 04.976 133 29.884 IPS3 200 3,1 1 31 35
9 2002 02IPS4-2 BMH02-9 70 04.980 133 29.983 IPS4 895 3,1 1 31 35
10 2003 03IPS4-2 BMH03-10 70 53.761 129 45.721 IPS4 895 4,3 1 30 33

Table 12 describes the types of sonars used.

Table 12. Sonar Type Descriptions
Sonar Type Description
IPS1 IPS Model 1
IPS2 IPS Model 2
IPS3 IPS Model 3
IPS4 IPS Model 4
NB-ADCP Narrow-band ADCP
WASP Self-contained echo sounder

Table 13 lists the mooring type codes used in Table 11.

Table 13. Mooring Type Codes
Mooring Code Description
0 Unconventional mooring
1 Stand-alone mooring for IPS
2 Stand-alone mooring for ADCP
3 Mooring supporting both IPS and ADCP

Location type names in Table 11 have the following convention and are described in Table 13.

PPPYY-X[X]

Where:

Table 13. Location Type Codes
Variable Description
PPP Project Acronym (over time this project has had different names)

ISC: Ice Subsurface Characterization project
ITT: Ice Type and Thickness project
ITC: Ice Thickness and Climate project
BMH: Beaufort Marine Hazards project
YY 2-digit year of when sonar was deployed
X[X] Site number: 1 - 10 (Note: some sites have an A and B location)


Mooring Types

A Type 1 mooring is a taut-line configuration that is as short as practical to minimize the hazard from drifting ice in shallow water (less than 40 m depth). See Figure 3. Five plastic floats (Viny 12B3) support the in-water weights of the instrument and the acoustic transponding releases (4 floats are sufficient for the smaller 420 kHz IPS). Viny floats provide an excellent buoyancy-to-drag ratio; the drag of this float is only six percent of the buoyancy with a 0.5 m/s current. With this design, the IPS maintains a zenithal orientation within ±2° and moves vertically by less than two centimeters in such a current. The instrument is contained within a 316 stainless steel frame, which provides attachment points for other devices and grappling points for under-ice recoveries. Some frames were fitted with a low-frequency radio beacon (Pieps 457) and a pinger, both of which only switch on following release in response to lower ambient pressure. The deactivation of these beacons at working depth prevents interference with the sonar and conserves battery power over long deployments. The mooring is equipped with two transponding releases, connected in parallel for redundancy. The anchor weight is built from clumped chain, permitting the weight used to be no more than necessary. The mooring is assembled on the ice or the deck of a ship and deployed anchor first by free-fall to the seabed.

Type 1 Mooring
Figure 3. Type 1 Mooring (Click for a high resolution version)

A Type 2 mooring is very similar to a Type 1 mooring. See Figure 4. Four plastic floats are sufficient, since the zenith-pointing constraints on the ADCP are much less severe than those for the IPS (±5° in 0.5 m/s current). The principal differences are a vane on the instrument frame and a swivel beneath it. By aligning with the flow, the vane reduces azimuthal oscillations of the instrument that are associated with vortex shedding. Azimuthal stability simplifies the correction of the time-averaged ADCP headings for compass non-linearity, since the necessary correction is well defined. This is not so if the ADCP rotates appreciably over the averaging interval. When Type 1 and Type 2 moorings were used at a site, they were placed no closer than the depth of water (50 to 80 m) to avoid acoustic cross-talk. In order for the two instruments to view the same ice as closely as is practical, the separation of the moorings has rarely exceeded 250 m.

Type 2 Mooring
Figure 4. Type 2 Mooring (Click for a high resolution version)

A Type 3 mooring is essentially a Type 1 mooring connected above a Type 2 mooring by approximately 50 m of line. See Figure 5. Since the IPS interferes with the ADCP operation if too close (less than 30 m), this mooring can be used in water deeper than 80 m. More flotation is needed if the depth of water exceeds 120 m. The IPS, which produces the best data if close to the ice, is best positioned deeper than 40 m to avoid impact with drifting ridge keels. A heavier anchor is required for the Type 3 mooring and deployment and recovery operations can be more complicated. The vertical stability of the IPS, which is important for the accurate calibration of ice draft, is degraded because the IPS may occasionally be pulled down as much as a meter by the stronger currents in this area. Vertical displacement of the sonar should be tracked by measuring pressure at more frequent intervals when using a Type 3 mooring, perhaps every two minutes or less.

Type 3 Mooring
Figure 5. Type 3 Mooring (Click for a high resolution version)
Background color on
Data Processing

For a complete description of the processing steps used to obtain this data, please see Draft and Movement of Pack Ice in the Beaufort Sea: A Time-Series Presentation April 1990 - August 1999 (H. Melling and D.A. Riedel 2004). Download this document without graphics: draft_movement_beaufort_2004.pdf (PDF, 1.6 MB). Download this document in zip format with graphics: draft_movement_beaufort_2004.zip (zip file, 4.7 MB).

Ice Draft Data

A brief summary of the processing steps required to obtain ice draft data from an IPS follows:

  1. Establish the relationships between travel time and range and between pressure and depth using observations made during sonar deployment and retrieval. These calculations establish the exact depth of the IPS, information that is essential to the calibration of ice draft.
  2. Determine the drift in calibration of the pressure sensor caused by creep during deployment. Derive a time-dependent calibration for pressure values recorded by the IPS.
  3. Determine the range correction factor, β(t), which is the ratio of the true target range to that calculated using the assumed depth-averaged sound speed. Preliminary values of range are calculated using an assumed average sound speed for the upper ocean (typically 1437 m/s).
  4. Verify that the temporal variation of the range correction factor is consistent with plausible seasonal changes in the temperature and salinity of waters between the sonar and the surface.
  5. Examine each target carefully as a potential ice-free surface suitable for use as a reference for zero draft.
  6. If the preliminary draft of a verified ice-free target is not zero, adjust β(t) at the time of observation to obtain a zero value. Reiterate the preceding two steps until results are consistent with the ±0.05 m target for accuracy. Since calibration points for β(t) are highly asynchronous, a reasonable interpolation between values must be achieved.
  7. Determine which portions of each record were obtained during prolonged ice-free conditions. Remove the ice-free (less than one tenth ice concentration) periods from the data files to facilitate further processing.
  8. Edit the ice draft record using automated methods and then manual review.
Challenges in Processing Ice Draft Data from IPS
  1. Identifying ice-free waters
  2. Loss of echoes
  3. Spurious echoes
  4. Wind waves in leads

Ice Velocity Data

A brief summary of the processing steps required to obtain ice velocity data from an ADCP follows:

  1. Edit the ice motion records obtained by the ADCP using automated methods and then manual review.
  2. Where justifiable, repair gaps in the ice motion record using the best available estimates of ice drift from other sources.
  3. Set the ice velocity identically equal to zero during intervals when the ice is stationary.
  4. Calibrate the time series of ice drift. The narrow band ADCP stores values that are calculated with an assumed sound speed of 1536 m/s, whereas the value relevant to ice drift speed, that at the ice water interface, is about 1437 m/s (water at 2 db pressure, 32 salinity and -1.5° C). The conversion factor is 0.935.
  5. Integrate ice velocity to a Eulerian displacement in UTM coordinates.
Challenges in Processing Ice Velocity Data from the ADCP
  1. Signal degradation in ice free waters
  2. Signal degradation under stationary smooth ice
Background color on
Error Sources

For a complete description of the possible error sources, please see Draft and Movement of Pack Ice in the Beaufort Sea: A Time-Series Presentation April 1990 - August 1999 (H. Melling and D.A. Riedel 2004). Download this document without graphics: draft_movement_beaufort_2004.pdf (PDF, 1.6 MB). Download this document in zip format with graphics: draft_movement_beaufort_2004.zip (zip file, 4.7 MB).

Ice Draft Data

The following are a list of possible sources of error when collecting ice draft data:

  • Relationship between pressure and depth:
    As density varies with time, the relationship between pressure and depth changes. Fortunately, the apparent depth change caused by the increase in density between summer and winter is negligible in this application where the IPS is at less than 100-m depth in Arctic waters (Melling and Riedel 1993).
     
  • Impact of seasonal change in the sound-speed profile:
    The impact of seasonal change in the sound-speed profile on the apparent range to the surface is not negligible. Melling and Riedel (1993) demonstrate that changing sound speed introduces an error of at least 0.2 m between winter and summer. Temperature and salinity profiles by Conductivity, Temperature, and Depth (CTD) are rarely available other than at deployment and recovery. The only feasible method to adjust ranges for secular change in sound speed is to use corrections derived from the apparent draft of open-water areas which pass through the sonar beam. These occasional tie points constrain the selection of a plausible curve that can be used to correct ranges at all times between deployment and recovery. Tie points must be selected with great care, since mistakes in selection introduce systematic errors (under estimates) in ice draft.
     
  • Sound speed varies with depth:
    Because sound speed varies with depth, an exact range correction will vary with the draft, or range, of the target as well as on the time of year. Thus, accuracy is best for thin-ice targets near the surface and deteriorates with increasing draft. In winter in the Beaufort Sea when open areas freeze very quickly, tie points for range correction are scarce. However, the temperature and salinity of water beneath continuous ice change only slowly from week to week. In summer, ice-free targets are viewed many times each day, but the temperature and salinity of the upper ocean are much more variable. River inflow, ice melt water, and solar heating are contributing factors. Our estimate for accuracy in the draft of level ice from moored sonar in this program is ±0.05 m.
     
  • Rough ice:
    Under rough ice, the first contact of the transmitted sound pulse with the ice may not be exactly overhead, because the sonar beam has finite angular width. Thus the local draft of ice will be overestimated on average in rough ice. The magnitude of the overestimate cannot be generally specified, since it depends not only on the beam pattern of the sonar, but on its source level and sensitivity and on the geometry, scattering cross-section and range of the ice. By reducing the field of view of the IPS to 0.8 m from a depth of 50 m (Melling 1998b), this systematic error has been reduced to a practical minimum.

Ice Velocity Data

The following are a list of possible sources of error in collecting ice velocity data.

  • Sampling error in the estimation of mean Doppler shift:
    The sampling error is random and a function of the acoustic frequency and pulse length used by the ADCP, the signal-to-noise ratio, and the number of pings averaged. For the ADCP configuration of these deployments, the sampling error was 0.7 cm/s in each component of velocity at good signal-to-noise ratio. When an ADCP is tracking floes which have a very smooth under surface, much of the transmitted pulse, incident at 30°, is reflected onward at 30°. A weak backscattered signal may be buried in noise at the receiver. At poor signal-to-noise ratio, the sampling error, estimated from the root-mean-square value of the vertical velocity component, increases to several cm per second. Poor precision in Doppler velocity is only problematic when a weak target is stationary over the IPS.
     
  • Uncertainty in the sound speed used to convert Doppler shift to velocity:
    The conversion of velocity components from a beam-referenced coordinate system to an Earth-referenced system is based on measurements of tilt and heading by the ADCP. The accuracy of the tilt measurement is not critical, since a (large) error of 5° in tilt generates an error of only 3% in a velocity component. Errors in the compass measurement of heading, however, translate directly into errors in the direction of ice drift. The non-linearity in the ADCP compass response can be quite large (±15°) in the Beaufort Sea because the horizontal component of the geomagnetic field is only 7000 nT. Corrections must be made.
     
  • Uncertainty in ADCP heading, pitch, and roll:
    Compasses were calibrated in a non-magnetic enclosure located near the Polar Continental Shelf Project (PCSP) base at Tuktoyaktuk. The output of the compass was measured at 11.25° increments in heading and corrections for the observed non-linearity were entered into a look-up table in the Erasable Programmable Read-Only Memory (EPROM) within the ADCP. The ADCP rotation was repeated to determine the residual non-linearity. For the compass manufactured by KVH Industries, the systematic sinusoidal component residual was about ±2° and the random component about the same magnitude. For the compass manufactured by EG&G (s/n 0179), the residual was much larger. Compass rotations were repeated at greatly different temperature and after intervals of 3-5 years, but no significant differences dependent on temperature or elapsed time were noted.

    Corrections for systematic residual non-linearity in heading are applied during data processing, leaving a random error of ±2°. To permit assignment of a unique value of heading correction for each ensemble (which represents data gathered over 15-45 minutes of pinging), each ADCP mooring was equipped with a vane and a swivel. The action of current on the vane stabilizes the heading of the ADCP during acquisition of an ensemble.
Background color on
Quality Assessment

The quality of the data has been established in two ways:

  1. A meticulous attention to detail in calibration and processing
  2. Published literature and the scientific use of data which has not turned up any obvious inconsistencies or enigmas and has generated new understanding of pack ice and Arctic change.

There are few independent means of assessing the accuracy of the ice-draft values in this set. Thickness measured via drill holes through the dominant first-year ice mode in late winter has been used as an occasional check on ice draft. The ice velocity data are occasionally compared with Lagrangian data provided by satellite-tracked ice-drift buoys that pass nearby. Generally, however, data quality is established by very onerous procedures in which the detection and identification of patches of truly ice-free water is critical. Such may be found only at intervals of several weeks during November through April. Strenuous precautions are taken to avoid misidentification of thin ice as open water, since this introduces bias that can seriously compromise the value of these data for climate study. Inter-comparison data indicate that the ±5 cm target accuracy for level ice is generally attained. Draft accuracy for rough ice of deeper draft is likely less good and quite variable; it may increase to ±30 cm at times.

Background color on

References and Related Publications

Contacts and Acknowledgments

H. Melling and D. A. Riedel
Science Branch - Pacific Region
Institute of Ocean Sciences
P.O. Box 6000, Sidney, BC
Canada V8L 4B2

Acknowledgments: 

Distribution of the data set from NSIDC is supported by funding from NOAA's National Environmental Satellite, Data, and Information Service (NESDIS) and the National Geophysical Data Center (NGDC).

Document Information

Document Authors

A. Windnagel and L. Ballagh compiled and created this document from correspondence with H. Melling and F. Fetterer and from Draft and Movement of Pack Ice in the Beaufort Sea: A Time-Series Presentation April 1990 - August 1999 (H. Melling and D.A. Riedel 2004). Download the Melling and Riedel 2004 document without graphics: draft_movement_beaufort_2004.pdf (PDF, 1.6 MB) or download it in zip format with graphics: draft_movement_beaufort_2004.zip (zip file, 4.7 MB).

Document Creation Date

September 2008

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