Get Data

DOWNLOADING DATA VIA HTTPS

To learn more about Earthdata Login and to register for an account, please see How to Register with Earthdata Login. Once you have logged in, data can be downloaded via a Web browser, command line, or client. For help with downloading data, please see Options Available for Bulk Downloading Data from HTTPS with Earthdata Login.

HTTPS

Data Set ID:
NSIDC-0393

Arctic Sea Ice Freeboard and Thickness, Version 1

This data set provides measurements of sea ice freeboard and sea ice thickness for the Arctic region. The data were derived from measurements made by from the Ice, Cloud, and land Elevation Satellite (ICESat) Geoscience Laser Altimeter System (GLAS) instrument, the Special Sensor Microwave/Imager (SSM/I), and climatologies of snow and drift of ice.

Geographic Coverage

Parameter(s):
  • Sea Ice > Ice Depth/Thickness
  • Sea Ice
Spatial Coverage:
  • N: 86, S: 65, E: 180, W: -180

Spatial Resolution:
  • 70 m x 70 m
Temporal Coverage:
  • 17 February 2008 to 21 March 2008
  • 20 May 2005 to 23 June 2005
  • 17 February 2005 to 24 March 2005
  • 3 October 2004 to 8 November 2004
  • 18 May 2004 to 21 June 2004
  • 17 February 2004 to 21 March 2004
  • 25 September 2003 to 19 November 2003
  • 20 February 2003 to 29 March 2003
  • 4 October 2008 to 19 October 2008
  • 21 October 2005 to 24 November 2005
  • 22 February 2006 to 27 March 2006
  • 24 May 2006 to 26 June 2006
  • 25 October 2006 to 27 November 2006
  • 12 March 2007 to 14 April 2007
  • 2 October 2007 to 5 November 2007
(updated 2014-04-15)
Temporal Resolution: 1 day
Data Format(s):
  • Binary
  • PNG
  • GeoTIFF
  • ASCII Text
Platform(s) DMSP, ICESAT
Sensor(s): GLAS, SSM/I
Version: V1
Data Contributor(s): Donghui Yi, H. Zwally
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.

Yi, D. and H. J. Zwally. 2009, updated 2014-04-15. Arctic Sea Ice Freeboard and Thickness, Version 1. [Indicate subset used]. Boulder, Colorado USA. NSIDC: National Snow and Ice Data Center. doi: http://dx.doi.org/10.5067/SXJVJ3A2XIZT. [Date Accessed].

Back to Top

Collapse All / Open All

Detailed Data Description

Format

ASCII

Each space delimited ASCII vector data file contains file headers followed by columns for latitude, longitude, freeboard, and thickness.

Longitude in WGS 84 degrees in the approximate range of 0 to 360. 
Freeboard in meters above local sea level at the time of the measurement. 
Thickness in meters.

Freeboard = 0 means the sea ice surface is at sea level. The average of the lowest 1percent reference elevation is used as reference sea level, so about 0.5 percent of the data are below sea level, that is, negative freeboard. Freeboard values less than zero are set to zero.

Note: Thickness values of -999 indicate that a thickness value could not be calculated and should be treated as missing data.

Binary

The freeboard and thickness binary gridded files are 304 columns x 448 rows containing little-endian floating-point values in meters. The grid is the 25 km polar stereographic grid used for SSM/I processing.

Background color on
File and Directory Structure

Data are available on the HTTPS site in the following directory:

https://daacdata.apps.nsidc.org/pub/DATASETS/NSIDC0393_GLAS_SI_Freeboard_v01

Background color on
File Naming Convention

ASCII File Names

ASCII files are named according to the following convention and as described in Table 1:

Example file name: 

laser3i1343001.txt
laserLPTTTTCCC.txt

Table 1.Valid Values for the ASCII File Name Variables
Variable Description
laserLP GLAS Laser Period used to collect data. LP = 3d, 3e, 3f, 3g, 3h, or 3i
TTTT Laser track number: 0001 through 0432, or 1282 through 1354
CCC Cycle of Reference Orbit: 001, 002, or 003
.txt Identifies the file as ASCII data

Binary File Names

Binary gridded files are named according to the following convention and as described in Table 2:

Example file name: laser3d_freeboard_mskd.img

Example file name: 

laser3d_freeboard_mskd.img.hdr
laserLP_nnnnnnnnn.mskd.zzz

Table 2. Valid Values for the Binary Grid File Name Variables
Variable Description
laserLP GLAS Laser Period used to collect data. LP = 3d, 3e, 3f, 3g, 3h, or 3i
_nnnnnnnnn Indicating 'freeboard' or 'thickness'
_mskd Indicating masked image
.zzz Identifies file type: .img (image), .img.hdr (header), or .png (Portable Network Graphics)

Grid Cell Center Latitude and Longitude Files

The grid cell center files provide the latitude and longitude for the center of each grid cell as a little-endian floating-point number. The files are named according to the following convention, and as described in Table 3.

Example file name: 

PS25km_north_lon.img
PS25km_north_mmm.zzz

Table 3. Valid Values for the Grid Cell Center File Name Variables
Variable Description
PS Polar Stereographic
25km 25 kilometer grid
_north northern hemisphere
mmm latitude (lat) or longitude (lon)
.zzz Identifies file type: .img (image), img.hdr (header)

Mask File

The mask file consists of values 0 = water, 1 = land. The file is named gsfc_25n.msk. The corresponding header file is gsfc_25n.msk.hdr.

Background color on
File Size

ASCII File Size

ASCII file sizes range from 2 K to 1 MB

Binary File Size

The .hdr files are each 749 bytes. 
The .img files are each 532 KB.
The .png files range from 31 KB to 44 KB.

Background color on
Volume

ASCII

The total volume for the ASCII track data files is approximately 932 megabytes.

Binary

The total volume for the binary gridded image, header and masked image files is approximately eight megabytes.

Background color on
Spatial Coverage

Arctic region

Southernmost Latitude: 65° N
Northernmost Latitude: 86° N
Westernmost Longitude: 180° W
Easternmost Longitude: 180° E

Note: The ASCII data are reported with longitude of 0 degrees to 360 degrees.

Spatial Resolution

The resolution of the gridded data is 25 km. Vectors have a resolution of about 170 meters in the along-track direction, or the distance between the centers of adjacent GLAS footprints. Several kilometers typically separate the ICESat GLAS data tracks.

Projection and Grid Description

Two geoid models are used in this study. North of 64° N, the Arctic Gravity Project (ArcGP) latitude limit, the ArcGP geoid is used. The Earth Gravitational Model 1996 (EGM96) geoid is used below 64° N. The difference, D, between ICESat measured sea level and the geoid was used to create a 5 km D-grid for each period. The mean of the 5 km D-grids was used as an improved geoid.

Background color on
Temporal Coverage

Release 28 Laser Identifier 1AB: 2003-02-20 to 2003-03-29
Release 28 Laser Identifier 2A: 2003-09-25 to 2003-11-19
Release 28 Laser Identifier 2B: 2004-02-17 to 2004-03-21
Release 28 Laser Identifier 2C: 2004-05-18 to 2004-06-21
Release 28 Laser Identifier 3A: 2004-10-03 to 2004-11-08
Release 28 Laser Identifier 3B: 2005-02-17 to 2005-03-24
Release 28 Laser Identifier 3C: 2005-05-20 to 2005-06-23
Release 28 Laser Identifier 3D: 2005-10-21 to 2005-11-24
Release 28 Laser Identifier 3E: 2006-02-22 to 2006-03-27
Release 28 Laser Identifier 3F: 2006-05-24 to 2006-06-26
Release 28 Laser Identifier 3G: 2006-10-25 to 2006-11-27
Release 28 Laser Identifier 3H: 2007-03-12 to 2007-04-14
Release 28 Laser Identifier 3I: 2007-10-02 to 2007-11-05
Release 28 Laser Identifier 3J: 2008-02-17 to 2008-03-21
Release 28 Laser Identifier 3K: 2008-10-04 to 2008-10-19

Release Schedule lists the temporal coverage of each ICESat/GLAS product.

Temporal Resolution

Each ICESat campaign period represents one Arctic-wide ice thickness assessment. Thus the temporal resolution for this data set is approximately two to three times per year.

Background color on
Parameter or Variable

The parameters of this data set are sea ice freeboard, and sea ice thickness.

Parameter Description

Sea ice freeboard is the height in meters of the sea ice above the water level. Thickness is the thickness in meters of the sea ice.

In the ASCII files, the thickness value -999 indicates missing values, possibly due to failure of the thickness algorithm for certain corresponding freeboard measurements.

The binary gridded *.img and *.png files have been masked. Table 4 shows the mask color value assignments to surface features.

Table 4. Binary Image File and PNG File Values
.img File Value .png File Value Surface Feature
-4 gray = 128 land south of 65° N
-3 gray = 164 land at or north of 65° N
-2 gray = 64 water south of 65° N
-1 gray = 96 water at or north of 65° N
greater than or equal to 0 full color (rainbow) applied across parameter value range sea ice freeboard or thickness measurements in meters
 

Sample Data Record

ASCII

The following sample of the laser3d0001002.txt ASCII file shows header information, and the first four records of Latitude, Longitude, Freeboard and Thickness values.

Binary

The following sample shows the laser3d_freeboard_mskd.img masked binary gridded image.

 laser3d_freeboard.img binary gridded image

The following sample shows the laser3d_freeboard_mskd.png masked binary gridded image.

 laser3d_freeboard_masked.jpg masked binary gridded image

Background color on

Software and Tools

Software and Tools

ASCII

The ASCII track data may be displayed with any Web browser or plain text display software.

Binary

The binary gridded data may be displayed using ENVI, ArcGIS, or other similar software packages.

Background color on
Quality Assessment

For further discussion on quality assessment of GLAS products, refer to the Quality Assessment section of the NSIDC GLAS/ICESat L1 and L2 Global Altimetry Data Web page.

Background color on

Data Acquisition and Processing

Theory of Measurements

Two geoid models are used in this study. The ArcGP geoid is used north of 64° N since that is the ArcGP latitude limit. The EGM96 geoid is used below 64° N.

Elevations varying more than plus-or-minus 4 meters are not used. This condition filters out some land, island, and iceberg data. These conditions are the same as in Zwally et al. 2008.

ICESat measures a surface elevation profile referenced to an ellipsoid. Due to the limited accuracy of the geoids and ocean tide models, and poor knowledge of the dynamic topography, sea-ice surface elevation referenced to a geoid cannot be regarded as sea-ice freeboard. The information needed to calculate sea-ice freeboard is the elevation difference between the top of the snow surface, local sea levels, and snow height and density above the snow/ice interface. If the elevation difference is known, even if the absolute elevations are biased, the sea-ice freeboard can be determined. Thus, the knowledge of relative elevation is crucial while absolute elevation is less important. This is the underlying concept in the derivation of freeboard.

In this study, constant densities of ρW = 1023.9 kg m-3 and ρI = 915.1 kg m-3 are used to calculate sea ice thickness from the freeboard. There is no spatial variation of now density ρS. Snow density, including the time variation, is based on Kwok 2008. The range of the snow density is 0.16 to 0.40.

Refer to the Processing Steps section below for details on the application of theory of measurements.

Background color on
Data Acquisition Methods

Freeboard is measured from ICESat elevation profiles (Zwally et al. 2008). Snow depth is interpolated, both spatially and temporally, from climatology snow depth in situ measurements (Warren et al. 1999). Thickness is estimated from ICESat freeboard and climatology snow depth. SSM/I daily ice concentration data (Gloersen et al. 1992) from January 2003 to October 2008 are used to determine sea ice boundaries.

Background color on
Derivation Techniques and Algorithms

The grids of sea ice freeboard and thickness were derived from the corresponding ICESat point measurements of freeboard and thickness using the following steps:

  1. Freeboard/thickness values are estimated for each valid ICESat sea ice observation during the laser period.
     
  2. The latitude/longitude of each freeboard/thickness observation is mapped into a column/row for a particular cell of the 25 km resolution 304 by 448 Polar Stereographic grid. 
     
  3. All the freeboard/thickness observations that fall into the same Polar Stereographic grid cell are averaged together resulting in a single freeboard/thickness value pair for each cell. This resampling algorithm is commonly referred to as "drop in the bucket" resampling. The cell will be empty if there is no data in that cell.
     
  4. The daily SSM/I Goddard Space Flight Center (GSFC) sea ice concentration values for each cell in the Polar Stereographic grid for the laser periods are averaged together for each ICESat campaign. Those cells that do not have a sea ice concentration value, which are cells that fall within the pole hole poleward of 84.5° N, are assigned values. Values from cells across eight equiangular sectors, nearest to the estimation point, are formed into a weighted average. The weight, w, is based on distance, d, from the estimation point to the nearest cell value in a particular sector, via the following relation in Equation 1 which is described in Table 6:
     
    Pulse Broadening Parameter (Equation 1)

    Where:
     
    Table 6. Cell Value Weighted Average Equation Description
    Variable Description
    w Weight
    d Distance
    r Maximum radius from which to draw samples. To interpolate values across the polar region, r was set to 420 km.

     
  5. Assign 0 to freeboard/thickness grid cells where mean ice concentration is less than 20 percent. For freeboard/thickness grid cells having no value and ice concentration greater than or equal to 20 percent, interpolate values.

Processing Steps

Sea ice freeboard and thickness are calculated from ICESat GLAS ground tracks using the following steps.

  1. Elevation data filtering. Before calculating the sea-ice freeboard and thickness, the following conditions were applied to filter out data contaminated by clouds, saturation, and land or islands. 
     
    1. Gain limit. An upper limit of detector gain is applied to filter out stronger atmospheric attenuated waveforms. The upper limit for the laser periods are: 50 counts for L1, L2, L2A,L2B,L3A, and L3B; 80 counts for L3C,L3D,L3E,L3F,L3G,L3H, and L3I; 120 counts for L2C,L3J, and L3K (Yi et al. 2011).
       
    2. Pulse broadening limit. Define a pulse-broadening parameter, S. See Equation 2 which is described in Table 7. 
       
      Pulse Broadening Parameter (Equation 2)

      Where:
       
      Table 7. Pulse-Broadening Parameter Equation Description
      Variable Description
      S Measure of the broadening of the transmitted pulse associated with surface topography and the undesirable effects of saturation and atmospheric forward scattering
      c Speed of light
      σR Echo waveform 1-sigma pulse width
      σT Transmitted waveform 1-sigma pulse width

      Heavily saturated waveforms and forward scattering waveforms have broadened pulse widths, so data with S larger than 0.8 m are discarded. 
       
    3. Reflectivity limit. Heavily saturated waveforms also tend to have very high apparent reflectivity, and forward scattering waveforms tend to have low reflectivity. Therefore, data with reflectivity smaller than 0.05 or larger than 0.9 are discarded. 
       
    4. Elevation limit. Elevation varying more than plus or minus four meters are not used. This condition filters out some land, island, and iceberg data. These conditions are the same as in Zwally et al. 2008.
       
  2. Geoid. Two geoid models are used in this study. The ArcGP geoid is used north of 64° N since that is the ArcGP latitude limit. The EGM96 is used below 64° N. 

    The difference, D, between ICESat measured sea level and the geoid was used to create a 5 km D-grid for each period. The mean of the 5 km D-grids was used as an improved geoid.
     
  3. Saturation correction and Inverse Barometer correction are applied as shown in Equation 3 and explained in Table 8. Both corrections are the same as in Zwally et al. 2008. Sea surface response to atmospheric pressure loading, the inverse barometer effect, is computed using the method described in the Aviso and Physical Oceanography Distributed Active Archive Center (PODAAC) User Handbook (Picot et al. 2003).
     
    ΔHib = 9.948 × (Patm - P) (Equation 3)

    Where:
     
    Table 8. Inverse Barometer Effect Equation Description
    Variable Description
    ΔHib Inverse barometer correction
    Patm Surface atmospheric pressure
    P Time varying mean of the global surface atmospheric pressure over the oceans

    ΔHib is applied to the ICESat elevation at the same time as the saturation correction. The surface atmospheric pressures used here are from the National Center for Environmental Protection (NCEP) (Stackpole 1994). The mean global surface atmospheric pressures over the ocean are from Collecte Localisation Satellites (Dorandeu and Le Traon 1999).
     
  4. SSMI daily ice con. To avoid open ocean in low ice concentration areas, freeboard = 0 is assigned to areas where ice concentration less than 20 percent. This is an empirical limit to balance the sea ice filtered out and the freeboard contamination introduced by open ocean water waves.
     
  5. Elevation. The ICESat measured surface elevation, Hie, the i_elev in product GLA06, is referenced to the TOPEX/POSEIDON ellipsoid. ICESat surface elevations have instrument corrections, dry and wet troposphere corrections, and tidal corrections applied. Elevation, h, is defined in Equation 4 and described in Table 9. 
     
    h = Hie + ΔHib + ΔHsat - hg (Equation 4)

    Where:
     
    Table 9. Elevation Equation Description
    Variable Description
    h Elevation above the geoid
    Hie ICESat measured surface elevation
    ΔHib Inverse barometer correction
    ΔHsat Saturation correction
    hg Geoid height
     
  6. Calculate 50 km running mean (hm) of elevation h. ICESat measures a surface elevation profile referenced to an ellipsoid. Due to the limited accuracy of the geoids and ocean tide models, and poor knowledge of the dynamic topography, sea-ice surface elevation referenced to a geoid cannot be regarded as sea-ice freeboard. The information needed to calculate sea-ice freeboard is the elevation difference between the top of the snow surface on the sea ice and local sea levels. If the elevation difference is known, even if the absolute elevations are biased, the sea-ice freeboard can be determined. Thus, the knowledge of relative elevation is crucial while absolute elevation is less important. Here we describe an algorithm to determine relative elevation and use this relative elevation to estimate sea-ice freeboard. By determining local ocean level and using only the relative elevation, the influences of the longer wavelength (greater than 50 km) factors such as geoid error, long wavelength laser pointing error and tidal error, which affect the absolute elevation, are removed from the freeboard calculation. 
     
  7. Calculate relative elevation: The relative elevation is calculated using Equation 5 and as described in Table 10.
     
    hr = h - hm (Equation 5)

    Where:
     
    Table 10. Relative Elevation Equation Description
    Variable Description
    hr Relative elevation
    h Elevation above geoid
    hm 50 km running mean of elevation h
  8. Sea level, hs, at any given point is determined by averaging the lowest one percent of the hr values within 50 km of that point. The one percent value was selected empirically. It provides enough points in calculating mean sea level to reduce measurement noise, and also minimizes the influence of thinner ice on the calculation. This value may be optimized further as we learn more about the distribution of leads in the Arctic. In extreme cases when there is no open water within the 100-km range, hs will measure the height of thin ice, thus underestimating freeboard. 
     
  9. hd(D). The difference between ICESat measured sea level and the geoid used is shown in Equation 6 as described in Table 11.
     
    hd = hm - hs (Equation 6)

    Where:
     
    Table 11. Difference Between ICESat Measured Sea Level and the Geoid Used Equation Description
    Variable Description
    hd Difference between the ICESat measured sea level and the geoid used
    hm 50 km running mean of elevation h
    hs Sea level
  10. Freeboard height, F, at a given point is calculated as shown in Equation 7 and described in Table 12.
     
    F = hr - hs (Equation 7)

    Where:
     
    Table 12. Freeboard Height Equation Description
    Variable Description
    F Freeboard height
    hr Relative elevation
    hs Sea level

    To have a valid F at a point, there must be enough valid elevation measurements available within 50 km of that point. In this study, a point is discarded if less than 50 percent (300 points) of the total 600 points are available.
     
  11. Sea-ice thickness, according to Archimedes buoyancy principle, is shown in Equation 8 and described in Table 13. 
     
    T = ρW F - ρW - ρS TS (Equation 8)


    ρW - ρI ρW - ρI

    Where:
     
    Table 13. Sea-ice Thickness Equation Description
    Variable Description
    T Sea ice thickness
    F Freeboard height
    TS Snow depth
    ρW Water density
    ρS Snow density
    ρI Sea ice density

    In this study, constant densities of ρW = 1023.9 kg m-3 and ρI = 915.1 kg m-3 are used to calculate sea ice thickness from the freeboard. There is no spatial variation of ρS. Snow density, including time variation, is based on Kwok 2008.

    Since snow depths and densities available for converting freeboard to thickness are at 25 by 25 km grid scale, these gridded values reflect grid scale mean snow depth and density over the arctic but do not have the small scale variation that can be directly used to do the ICESat shot to shot freeboard/thickness conversion. Snow depth and density must be modeled to apply them to the individual measurements. The following method is used to do the freeboard/thickness conversion at a point.

    Due to the dynamic nature of arctic sea ice, snow falls are not always accumulated on the top of sea ice. They may fall on leads or open water. So the snow depth varies from one footprint to another. Here a snow accumulation factor was defined as Fx (= 0.4, 0.4, 0.6, 0.1 for FM, MA, MJ, and ON), shown below with dates and lasers: 

    Fx = 0.4 February to March (FM) Laser 3E
    Fx = 0.4 March to April (MA) Laser 3H
    Fx = 0.6 May to June (MJ) Laser 3F
    Fx = 0.1 October to November (ON) Laser 3D, 3G, 3I

    It was assumed when freeboard is larger than Fx, snow is fully accumulated on sea ice; when freeboard is less than Fx, snow accumulation on sea ice is proportional to the ratio δ=Fi/Fx. The following five conditions are applied when converting Fi to Ti: 

    < >If Fi < 0, set Fi = 0if Fi < Fx, set δ=Fi/Fx and if Fi ³ Fx, set δ=1Ts=δ Ts', Ts' is snow depth of 25×25 km grids, Ts is snow depth used in the conversionif Ts > Fi, set Ts=Fiif Ci < 20%, set Fi=0,  Ci is bilaterally interpolated ice concentration at a point from SSM/I ice concentration Table 14. Snow Accumulation Factor DescriptionVariableDescriptionFifreeboardFxsnow accumulation factordratio Fi/FxTssnow depthCibilaterally interpolated ice concentration at a point from SSMI ice concentrationTisea ice thickness

Error Sources

Table 5 summarizes the GLAS single-shot error budget for elevation measurements (Zwally et al. 2002).

Table 5. GLAS Single-shot Error Budget for ICESat Elevation Measurements
Error Source Error Limit
Precision orbit determination (POD) 5 cm
Precision attitude determination (PAD) 7.5 cm
Atmospheric delay 2 cm
Atmospheric forward scattering 2 cm
Other (tides, etc.) 1 cm
RSS 13.8 cm

For further discussion on error sources, see Zwally et al. 2002, and also refer to the Error Sources section of the NSIDC GLAS/ICESat L1 and L2 Global Altimetry Data Web page.

Background color on
Sensor or Instrument Description

The data for this data set were acquired with the GLAS instrument onboard the Ice, Cloud, and land Elevation satellite (ICESat), from the SSM/I instrument on board the Defense Meteorological Satellite Program (DMSP), and from weather stations on the sea ice.

Background color on

References and Related Publications

Contacts and Acknowledgments

Donghui Yi
Stinger Ghaffarian Technologies, Inc. 
Cryospheric Sciences Branch
NASA Goddard Space Flight Center
Greenbelt, MD 20771 USA

Document Information

DOCUMENT CREATION DATE

September 2010

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