Sea Ice MEaSUREs – Data Products

Now GIS Compatible!

The sea-ice data available below draw on more than 11 years of nearly uninterrupted, three-day radar snapshots of sea ice from the RADARSAT-1 satellite (1995-2012) or four years of Envisat data (2008-2012) of the Arctic and Southern Oceans. Small-scale kinematics and deformation data are processed by tracking sea ice on a high-resolution grid. Some original RADARSAT-1 images are also available from Vertex at no cost to approved users.

Eulerian data in the Sea Ice MEaSUREs South collection falls within the red bounding box visible in this image.

Eulerian data in the Sea Ice MEaSUREs North collection falls within the red bounding box visible in this image.

Data Terms Explained


The Lagrangian sea-ice data products contain monthly measurements of dynamic and kinematic parameters over the Arctic Ocean sea-ice cover. “Lagrangian” refers to a mathematical way to study ice dynamics by noting changes in position and velocity of points over time. The sea-ice analysis in the dataset available here is initialized by laying a 10km-by-10km grid over a set of RADARSAT images of the sea ice during an initial 3-day period of a season. The grid forms cells for which a number of properties can be derived that describe the sea-ice dynamics. The grid points of the sea ice and related cell properties are then tracked throughout the season.

Four data products result from these measurements:

This image illustrates sea-ice-motion magnitude, one of the
parameters in the 
Lagrangian products, at three points in time in the Arctic Ocean basin. Dark blue indicates the most motion, more than 30 km a day. The spatial extent of the data illustrated here is the typical extent of the winter products.

1) Ice Motion – a record of the time and location of each point within the initial grid as tracked on RADARSAT images at approximate 3-day intervals. Note that a very small fraction of the points may be lost during the season through advection out of the Arctic basin, loss of ice, or untrackability of the ice cover.

2) Deformation – a record of the divergence, vorticity and shear occurring within each cell. As the verticies of each cell move within a time step, the kinematic properties can be calculated to characterize the response of the ice cover to stresses induced by wind and ocean currents.

3) Ice Age and Thickness – a record of the area, age and thickness of new ice and ridged ice that result from cell area changes. A spatial and temporal distribution of these ice areas is kept for each cell. If the area of a cell increases within a time step, an area of new ice is created. The ice thickness within all the new ice areas created during the season is increased using air temperature information. When the cell area decreases, the thinnest ice areas are rafted or ridged, depending on their thickness. The thickness of areas of ridged ice are also grown in time.

4) Backscatter Histogram – a record of the radiometric properties of the ice within each cell. A histogram of the radar brightness is kept within each cell at each time step, allowing the user to deduce multi-year ice fractions of the ice cover.

Note that in earlier winter products we use a 10km-resolution grid over the existing ice cover beginning sometime near the beginning of November. In later products we use a hybrid grid of 10km resolution over the multi-year ice pack and a 20km grid over the seasonal ice and began our analysis in early December so as to permit the tracking of the seasonal ice regions.

3-Day Gridded

The 3-day gridded datasets are produced from the Lagrangian products. Parameters from the Lagrangian dataset are processed to produce fields with constant grid spacing. The gridded parameters are ice age, ice thickness, backscatter histogram, divergence, vorticity, and shear. Data values cover a 3-day period on a 12.5km-x-12.5-km grid. Each of the downloadable files contains one month of these 3-day product files.

Melt Onset

The melt onset product consists of a gridded field containing the date of surface melt at each grid location. This date is derived from changes in the radar backscatter signature within the Lagrangian cells between April and June. The grid is at 10km resolution within the interior of the Arctic basin and 25km resolution near the coasts.


Between 2008 and 2012, an archive of Envisat SAR imagery of the Arctic and Southern Oceans was created. In a project, conducted by Senior Research Scientist Ron Kwok of the Jet Propulsion Laboratory, the SAR imagery from Envisat was used to produce a high-resolution dataset of small-scale sea ice kinematics and deformation. This work is supported under NASA’s Making Earth System data records for Use in Research Environments (MEaSUREs) program. The motion field is sampled in Eulerian mode. 

The objectives are to:

  • Process the Envisat data stream to construct ESDRs of small-scale ice motion of the Arctic and Southern Oceans;
  • Develop products that take advantage of the temporal (daily observations) and spatial sampling scheme of the Envisat mission;
  • Produce mosaics of the Envisat images of the Arctic Ocean and Southern Ocean.