Polar sea ice is both an indicator and a driver of global climate change. We now have over 40 years of satellite data (since the late 1970s) to directly monitor its evolution in concentration, area, and extent and almost 30 years (since the early 1990s) for its thickness.

In the Arctic, sea ice extent and volume have decayed in all seasons, with strongest reduction in late summer. This leads to a younger and more mobile sea ice cover in the Arctic Ocean. In the Southern Hemisphere, sea ice extent increased rather steadily until 2015, with much smaller coverage since 2016. The Special Report for a 1.5C Global Warming (IPCC SR15) documents that to limit global warming to 1.5C above pre-industrial levels could “substantially” reduce the risk of sea ice-free summers in the Arctic with respect to a 2C warming.

Although satellite observations are the foundation for most of our knowledge about the evolution of the global sea ice cover, progress is still acutely needed to improve the observations of the Sea Ice Essential Climate Variable, in particular to achieve better spatial resolution, better consistency across satellite missions and longer time series.


The ultimate objective of Sea Ice_cci is to advance the retrieval capability for two main variables of the Sea Ice Essential Climate Variable (ECV): Sea ice concentration and sea ice thickness. Our research plan was designed to approach the GCOS requirements and maximize the impact of CCI research to operational climate information services.

For sea ice concentration (SIC), our efforts will be towards improving the spatial resolution of the Climate Data Records, by using the high-frequency channels available on the microwave radiometry missions since 1992. Starting with SSM/I F10, we now have an almost 30-year time series with frequencies around Sea Ice CC90 GHz. They offer higher spatial resolution, but also lead to larger retrieval uncertainties compared to the “classic” frequency channels available since the late 1970s. We will develop innovative algorithms that will combine all frequency channels and balance between higher spatial resolution and lower accuracy. In addition, we will attempt to retrieve sea ice cover information from the legacy ESMR instrument on board Nimbus-5 (1972-1977). ESMR was a pioneering instrument and dedicated algorithms must be developed. All our SIC developments are coordinated with those of the EUMETSAT Ocean and Sea Ice Satellite Application Facility (OSI SAF, osisaf.met.no).

For sea ice thickness (SIT), one of our main objectives is to consistently extend the existing Envisat+Cryosat2 time series (2002-present) with data from ERS-1 and ERS-2 missions (1993 onwards). This is a challenging task due to the difference in technology between ERS and Envisat (ERS recorded blurred, lower-resolution pulses wrt Envisat). Another challenge is our incomplete knowledge of the snow cover since the early 1990s. Indeed, snow plays a crucial role in observing sea ice thickness using space-borne altimeters, both for radar penetration, and conversion from freeboard to thickness. Other general algorithm developments, such as better uncertainty characterization, and gap-filling of the large polar observation hole (north for 81N for most of the period) will be tackled. All our SIT developments are coordinated with those of the Copernicus Climate Change Service (C3S).

About the project

The objectives described above are ambitious, and will not be addressed up-front. We rather designed a series of three 1-year development cycles to gradually develop the algorithms, and prepare better climate data records.

Each yearly phase will include the following steps:

  1. Update user requirements
  2. Algorithm developments
  3. Prototype software and process data
  4. Scientific validation
  5. Climate assessment

A series of deliverables (documents and datasets) will be released throughout the project.

Sea Ice CCI now enters a 3rd project phase, the first under the "CCI+" umbrella. Previous two phases involved many of the same partners, and was led by S. Sandven (NERSC, NO).