CryoSat-2¶

AWI CryoSat-2 Sea Ice Thickness (version 2.6) and other Arctic sea ice thickness products from the CryoSat-2 radar altimeter

Available

At a glance¶

Note on period: The v2.6 record starts on 2010-10-21. Northern hemisphere thickness is produced only for the freezing season, October to April; May to September is empty, because surface melt prevents a reliable freeboard retrieval. "Present" therefore means up to the most recent winter.

Overview¶

Arctic winter sea ice thickness derived from the SIRAL radar altimeter on ESA's CryoSat-2: the sea ice freeboard (height above the water line) is measured, and thickness is computed under hydrostatic equilibrium using assumed snow depth and snow, ice and sea-water densities. This page uses AWI's v2.6, produced with pysiral, as the default. Besides thickness, each file contains radar and sea ice freeboard, the snow depth and densities used, sea ice draft and quality flags, which makes it easy to trace sources of error. ESA CCI, CPOM and ESA Cryo-TEMPO also produce products from the same source data, and their values differ with the processing.

Available variables¶

Variables in the AWI v2.6 monthly and weekly gridded (L3C) files, checked against the v2.6 output definitions in AWI's processing code (pysiral); full descriptions are in the Product User Guide. Each physical quantity comes with an uncertainty variable suffixed _uncertainty (e.g. sea_ice_freeboard_uncertainty).

instantaneous mean accumulated maximum minimum static — instantaneous: value at the time stamp · mean/accumulated/maximum/minimum: over the indicated interval · static: does not change with time

Thickness and freeboard¶

Auxiliary fields used in the thickness retrieval¶

Quality and status¶

The full variable list is in the official documentation.

How to get the data¶

• AWI FTP — CryoSat-2 v2.6 (L2P trajectory, L3C weekly and monthly grids) — Anonymous FTP (FTP client required) No login needed
  • Northern hemisphere data are in the nh/l3c_grid/month, nh/l3c_grid/isoweek and nh/l2p_trajectory folders. The Latest folders hold only the newest near-real-time files, not the full record, so use the reprocessed data for analysis.
• AWI SIRAL wiki — Sea Ice Thickness from Satellite Radar Altimetry — Web documentation (product description, operational news, document links)
  • The same page also covers the Sentinel-3A/B sea ice thickness produced with the same v2.6 processing.
• ESA Sea Ice CCI — CryoSat-2 northern hemisphere monthly thickness L3C v4.0 (CEDA) — Web download / HTTP directory No login needed
  • A climate-record reprocessing for 2010-10 to 2024-04, from the same processing lineage as AWI. ESA CCI sea ice terms of use apply.
• CPOM (Centre for Polar Observation and Modelling, UK) Sea Ice Thickness portal — Web download (NetCDF, CSV time series)
  • Based on CryoSat-2 and Sentinel-3, on a sparse 5 km polar stereographic grid (EPSG:3411). Provides near-real-time (last 2, 14 and 28 days) and monthly data, plus regional volume and mass time series.
• ESA Cryo-TEMPO sea ice products (Winter Sea Ice, Summer Sea Ice, Snow Depth on Sea Ice) — CS2EO platform / ESA MAAP (EO Sign In account required)
  • ESA's own thematic CryoSat-2 products. A separate summer (May–September) radar freeboard product is also available.
• C3S Climate Data Store — Sea ice thickness gridded data for the Arctic — Web download / CDS API
  • Requires an ECMWF account. Provides a long-term L3 multi-altimeter record and a gap-filled L4 product combining CryoSat-2, Sentinel-3 and SMOS.

Example code¶

Assumes a monthly gridded file downloaded from the AWI FTP. File names follow awi-siral-l3c-sithick-cryosat2--nh_25km_ease2--fv2p6.nc.

import glob
import xarray as xr

f = glob.glob("awi-siral-l3c-sithick-cryosat2-*-nh_25km_ease2-201903-fv2p6.nc")[0]
ds = xr.open_dataset(f)

sit = ds["sea_ice_thickness"]          # m
qf = ds["quality_flag"]                # 0 nominal, 1 intermediate, 2 low, 3 no data

# Keep only nominal and intermediate quality cells and compute the mean thickness
sit_ok = sit.where(qf <= 1)
print(float(sit_ok.mean()))

# Check how sensitive the thickness is to the assumed snow depth
print(float(ds["snow_depth"].where(qf <= 1).mean()))

Practical tips¶

• Thickness is not measured directly; it is converted from freeboard using assumed snow depth and densities. AWI v2.6 uses a monthly climatology that regionally merges the Warren climatology (halved over first-year ice) with AMSR2-based snow depth. If your results differ from another product, compare the snow_depth and sea_ice_density variables first.
• Arctic data exist only for October to April. For May–September you need a product made with a different method, such as the ESA Cryo-TEMPO Summer Sea Ice freeboard, and you should not splice it directly onto the winter data.
• AWI, CPOM and ESA CCI all start from the same CryoSat-2 data but differ in retracker, snow data and grid, so their mean thickness and trends differ. Use one product and one version within an analysis, and record the version you downloaded (e.g. v2.6).
• Because of its 92° orbit inclination, CryoSat-2 does not observe north of about 88°N. This area is marked satellite_pole_hole in status_flag, so decide how to handle it before computing pan-Arctic volume. Thin ice (below roughly 1 m) has large relative errors; the CS2SMOS product, which merges CryoSat-2 with SMOS, may be more suitable there.

Notices and change history¶

References for your paper¶

Dataset citation¶

AWI CryoSat-2 v2.6 Product User Guide (named by AWI in the files' references attribute) — DOI 10.5281/zenodo.10044554

@techreport{hendricks2023awi_cs2_v26,
  author      = {Hendricks, Stefan and Paul, Stephan},
  title       = {Product User Guide \& Algorithm Specification - {AWI} {CryoSat-2} Sea Ice Thickness (version 2.6)},
  institution = {Alfred-Wegener-Institut Helmholtz-Zentrum f{\"u}r Polar- und Meeresforschung},
  publisher   = {Zenodo},
  year        = {2023},
  doi         = {10.5281/zenodo.10044554}
}

ESA Sea Ice CCI CryoSat-2 monthly thickness v4.0 (CEDA; if you use this alternative)

Reference papers¶

AWI CryoSat-2 thickness retrieval method — DOI 10.5194/tc-8-1607-2014

@article{ricker2014cs2,
  author  = {Ricker, R. and Hendricks, S. and Helm, V. and Skourup, H. and Davidson, M.},
  title   = {Sensitivity of {CryoSat-2} {Arctic} sea-ice freeboard and thickness on radar-waveform interpretation},
  journal = {The Cryosphere},
  volume  = {8},
  number  = {4},
  pages   = {1607--1622},
  year    = {2014},
  doi     = {10.5194/tc-8-1607-2014}
}

Acknowledgements¶

AWI states that it uses CryoSat-2 Level-1 data (ESA), sea ice concentration and type (OSI SAF), mean sea surface (DTU21, DTU Space) and AMSR2 snow depth (Institute of Environmental Physics, University of Bremen) as inputs, so it is good practice to credit these sources in the acknowledgements. CC BY 4.0 requires attribution to AWI.

Information on the datasets used in this study was compiled with the help of the Korean Climate Data Navigator (KCDN) (Sim, 2026; https://doi.org/10.5281/zenodo.23300923).

Links¶

• AWI v2.6 Product User Guide (Zenodo)
• AWI CryoSat-2 version history
• pysiral (AWI processing code, GitHub)
• ESA CryoSat mission page
• CPOM — Celebrating 15 years of CryoSat-2 (mission extended until 2028)