Data Set ID:

CryoSat-2 Level-1B Waveforms, Sea Ice Elevation, and Surface Roughness, Version 1

This data set contains surface elevations from retracked CryoSat-2 waveforms, as well as model fitting parameters used to retrack the waveform. The primary data set used in the production of these data come from the ESA CryoSat-2 satellite.

This is the most recent version of these data.

  • Sea Ice > Sea Ice Elevation
  • Topography > Surface Roughness
Data Format(s):
  • NetCDF
Spatial Coverage:
N: 90, 
S: 55, 
E: 180, 
W: -180
Spatial Resolution:
  • 1650 m x 380 m
Temporal Coverage:
  • 15 September 2010
Temporal Resolution1 dayMetadata XML:View Metadata Record
Data Contributor(s):Nathan Kurtz, Jeremy Harbeck

Geographic Coverage

Other Access Options

Other Access Options


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.

Kurtz, N. and J. Harbeck. 2018. CryoSat-2 Level-1B Waveforms, Sea Ice Elevation, and Surface Roughness, Version 1. [Indicate subset used]. Boulder, Colorado USA. NASA National Snow and Ice Data Center Distributed Active Archive Center. doi: [Date Accessed].

Back to Top

Collapse All / Open All

Detailed Data Description

This data set contains surface elevations from retracked CryoSat-2 waveforms, geophysical corrections, as well as model fitting parameters used to retrack the surface elevation. Retracking is the process of selecting the point on the return radar waveform that corresponds to the mean scattering surface. Primary data sources are ESA's CryoSat-2 Level-1B Synthetic Aperture Radar (SAR) and SAR Interferometric (SARIn) data.

This data set provides information on Arctic sea ice thickness which is a key parameter for understanding changes in the climate. The data are needed for a variety of purposes ranging from model validation and initialization, data assimilation efforts, seasonal sea ice forecasting, and development of new retrieval methods.


The data files are in netCDF format (.nc). Each data file is paired with an associated XML file (.xml). The XML files contain file, date, location, platform, and instrument metadata.

Background color on
File and Directory Structure

Data are available via HTTPS in the following directory:

Within this directory, the folders are named for each year, month, and day of data collection, for example /2010.09.15/.

Background color on
File Naming Convention

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

Example File Names:

Refer to Table 1 for the valid values for the file name variables listed above.

Table 1. File Naming Convention
Variable Description
RDWES1B Short name for CryoSat-2 Level-1B Waveforms, Sea Ice Elevation, and Surface Roughness
MM The mission identifier. CS  = CryoSat (ESA-introduced variable)

File class, e.g.:

  • OFFL = Off-line Systematic Processing (ESA-introduced variable)
  • LTA_ = Long Term Archive
TTTTTTTTTT File type, e.g.:
  • SIR_SAR_1B: SIRAL SAR mode Level-1B
  • SIR_SIN_1B: SIRAL SIN mode Level-1B
yyyymmddThhmmss Start time window as extracted from job order (ESA-introduced variable)
YYYYMMDDTHHMMSS Stop time window as extracted from job order (ESA-introduced variable)
Cvvv Data reprocessing release C, followed by version number (ESA-introduced variable)

File format, e.g.:

  • NetCDF (.nc)
  • XML (.xml)
Background color on
File Size

As of April 2018, the total file volume is 25.2 GB. Since new data are added to this product on a daily basis, the total file volume will be updated accordingly in this document.

Background color on
Spatial Coverage

Spatial coverage for the NASA GSFC CryoSat-2 sea ice product currently corresponds to the sea ice-covered areas of the Arctic Ocean.

Southernmost Latitude: 55° N
Northernmost Latitude: 90° N
Westernmost Longitude: 180° W
Easternmost Longitude: 180° E

Spatial Resolution

Satellite along-track swath; footprint size: 380 m along-track by 1650 m cross-track

Projection and Grid Description

Geographic coordinates (latitude and longitude).

Background color on
Temporal Coverage

15 September 2010 to present.

Data are delivered on an ongoing, daily basis, with an approximately one-month delay from the current date. As summer data are not reliable for ice thickness due to retrieval issues caused by summer melt, data delivery began on 15 September 2010 and does not include data from 16 May to 14 September for any subsequent years.

Each data file covers a range of time from 1 second to greater than 2300 seconds, with an average length of ~175 seconds.

Temporal Resolution

Multiple footprints every second, with each file a portion of a satellite swath. There are between 1 and 130 files per day.

Background color on
Parameter or Variable

Scientific variables included in this data set are listed in Table 2 with the Parameter Origin NASA Kurtz or ESA.

Parameter Description

The data file contains fields as described in Table 2.

Table 2. Data Parameter Description
Parameter Description Units Parameter Origin
lat Latitude Degrees ESA
lon Longitude Degrees ESA
elev Elevation at center of range window Meters ESA
retrack_elev Elevation retracking correction Meters NASA Kurtz
roughness Surface roughness Meters NASA Kurtz
alpha Angular backscattering efficiency Dimensionless NASA Kurtz
norm_res Error of fit Dimensionless NASA Kurtz
peakiness Pulse Peakiness Dimensionless NASA Kurtz
stack_sd Stack standard deviation Dimensionless ESA
geophys_corr Sum of geophysical corrections Meters ESA
sat_alt Satellite Altitude Meters ESA
pitch Pitch Microradians ESA
roll Roll Microradians ESA
day Days since January 1st, 2000 Day ESA
sec Seconds since start of day Seconds ESA
i Row index within Cryosat file Index ESA
j Column index within Cryosat file Index ESA
amp_fit Amplitude fit parameter Dimensionless NASA Kurtz
time_shift Time shift Nanoseconds NASA Kurtz
wf_start_bin Beginning index of waveform subset Index NASA Kurtz
wf_end_bin Final index of waveform subset Index NASA Kurtz
phase Phase Microradians ESA

Elevation (elev)
Elevation at the center of range window with the geophysical corrections field and oscillator drift delay added in. Note that a correction for the lower propagation speed in snow is not applied to these data.

Elevation retracking correction (retrack_elev)
Waveform-fitting method-based retracking correction for the elevation field. This field must be added to the Elevation field to calculate the final retracked elevation.

Surface roughness (roughness)
Ice surface roughness is derived using the physical model to fit the CryoSat-2 waveform. The surface roughness is the standard deviation of the ice surface elevation with an assumed Gaussian height distribution.

Sum of geophysical corrections (geophys_corr)
Geophysical corrections for the wet and dry tropospheric delay time, ionospheric delay, dynamic atmospheric correction, ocean equilibrium tide, long period ocean tide, load tide, solid earth tide, and pole tide have been applied from the ESA CryoSat-2 L1B data products.

Retracking model fitting parameters (roughness, alpha, amp_fit, time_shift)
These parameters include the following variables to recreate the model waveform fit: roughness, alpha, amp_fit, and time_shift. The model which uses these parameters is described in Kurtz et al. (2014).

Error of fit (norm_res)
Defined as the sum of the squared difference of the normalized Cryosat-2 waveform and the model waveform fit.

Sample Data Record


Table 3. Sample Data [Entry 1 from above file]
Parameter Value Units
lat 67.97881 Degrees
lon 249.69783 Degrees
elev -27.227 Meters
retrack_elev -1.9908 Meters
roughness 0.1919 Meters
alpha 14.0 Dimensionless
norm_res 0.0794 Dimensionless
peakiness 0.0419 Dimensionless
stack_sd 13.26 Dimensionless
geophys_corr -2.211 Meters
sat_alt 724322.75 Meters
pitch -6.95E-1 Microradians
roll -0.001783 Microradians
day 6446.0 Day
sec 84050.0 Seconds
i 1 Index
j 44 Index
amp_fit 0.986 Dimensionless
time_shift -78.906 Nanoseconds
wf_start_bin 472 Index
wf_end_bin 599 Index
phase 626871.3 Microradians
Background color on

Software and Tools

The following external links provide access to software for reading and viewing HDF5 and netCDF data files. Please be sure to review instructions on installing and running the programs.

HDFView: Visual tool for browsing and editing HDF4 and HDF5 files.

Panoply NetCDF, HDF, and GRIB Data Viewer: Cross-platform application. Plots geo-gridded arrays from netCDF, HDF, and GRIB data sets.

For additional tools, see the HDF-EOS Tools and Information Center.

Background color on
Quality Assessment

Comparison of retrieved freeboards from CryoSat-2 with ATM airborne laser scanner data from Operation IceBridge suggest an uncertainty of 4-9 cm (Kurtz et al. 2014). A retracked elevation bias of ~20 cm has also been found in comparison to airborne laser scanner data from Operation IceBridge. This is anticipated to be corrected in a future version and does not affect the retrieval of sea ice freeboard, which is a relative measurement.

The norm_res parameter provides one measure of the elevation retrieval quality. This parameter is the sum of the squared difference between the normalized CryoSat-2 waveform and the modeled waveform fit. A lower norm_res indicates a better goodness of fit, and values higher than 0.5 are not used in the retrieval of freeboard in the CryoSat-2 Level-4 Sea Ice Elevation, Freeboard, and Thickness product.

Background color on

Data Acquisition and Processing

Background color on
Derivation Techniques and Algorithms

Processing Steps

Surface elevation and roughness are retrieved from individual CryoSat-2 waveforms by fitting a model waveform to the data. First, a look-up table for the CryoSat-2 impulse response convolved with the transmit pulse is produced as described in Kurtz et al. (2014) and placed on a regular grid. Each CryoSat-2 waveform is then sub-sampled to 128 range bins around the expected location of the surface return, with the start and end points of the waveform location provided in the wf_start_bin and wf_end_bin fields, and the waveform power is normalized to have a maximum value equal to 1. The waveforms are separated into expected lead returns and floe returns using the pulse peakiness and stack standard deviation parameters. Initial guesses for the model fit parameters and bounds are produced. The specification for the initial guesses and bounds are described in Kurtz et al. (2014), who used a 50% threshold tracker. However, the initial guess for the echo time shift of sea ice leads is now derived using a 70% threshold tracker. Once initial guesses and bounds are specified, the waveform fitting procedure is applied using the MATLAB lsqcurvefit function. If the waveform fit has an initial poor fit (defined as norm_res value greater than 0.3), then an iteration on the fitting is done using a range of values for the initial guess and bounds of the alpha parameter. If a lower value of norm_res is found in the iteration, then it is used in the retrieval process. The waveform fitting process produces the roughness, alpha, amp_fit, and time_shift parameters. The time_shift parameter is then converted into a range bin value and elevation retracking correction by finding the difference with the center of the range window.

Error Sources

The primary error source in the data comes from unmodeled physical parameters including backscatter from the snow surface and volume and surface roughness-induced backscatter variations within the footprint. For sea ice leads the dominant errors are due to the finite range resolution of the instrument and off-nadir lead returns.

Background color on
Sensor or Instrument Description

The ESA SIRAL instrument, the primary instrument on board CryoSat-2, is a radar altimeter which measures the surface elevation through knowledge of the spacecraft position and the time delay between the emission of the radar pulse and subsequent reflection from the surface.

The SIRAL instrument operates at a center frequency of 13.575 GHz and has a receiver bandwidth of 320 MHz. The SAR processing of CryoSat-2 utilizes an unfocused aperture synthesis technique which uses Doppler beam formation to reduce the footprint size in comparison with a beam-limited altimeter. The effective footprint size after post-processing is pulse-limited at 1650 m in the across-track direction and pulse-Doppler-limited to be 380 m in the along-track direction. The power-detected echoes contain 256 range bins in SAR mode and 1024 range bins in SARIn mode (Kurtz et al. 2014). The SAR mode is typically operated over sea-ice areas as well as ocean basins and coastal zones, whereas the SARIn mode is usually employed for the steep slopes of ice sheet margins, over small ice caps, and over mountain glacier regions.

Background color on

References and Related Publications

Contacts and Acknowledgments

Nathan Kurtz and Jeremy Harbeck
Greenbelt , MD 20771
Country: United States


We acknowledge support from NASA’s Cryospheric Sciences program from grant NNH12ZDA001N. We would also like to thank the European Space Agency for processing and providing the CryoSat-2 data.

Document Information


March 2018

No technical references available for this data set.

How To

Programmatically access data using spatial and temporal filters
The Common Metadata Repository (CMR) is a high-performance metadata system that provides search capabilities for data at NSIDC. A synchronous REST interface was developed which utilizes the CMR API, allowing you to ... read more