Ice classes are assigned from atmospherically corrected SSMIS brightness temperatures and ASCAT backscatter values, using a Bayesian approach. Both the sea ice edge product and the sea ice type product are classification products differing between different ice classes. Sea ice edge differs between the classes open water, open ice, and closed ice which are defined from the ice concentration thresholds of 30 proc. and 70 proc., respectively. The product series is operational since 2005 (version OSI-403-b since 15/09/2015, version OSI-403-c since 30/05/2017).
Other input :ECMWF forecast for atmospheric correction
Frequency :1 per day
Central time :12:00
Timeliness :5 h
Spatial coverage :global
Spatial sampling :10 km
File Formats :HL FTP Server (NetCDF4), EDC (NetCDF4), EUMETCast (NetCDF4)
Characteristics & methods :Discrimination Open ice/Closed ice/No ice. Multisensor analysis, daily average. Risk mitigation against sensor degradation.
Data Access
Main Diffusion Means :
HL FTP server, last 31 days (hosted by MET Norway) : ftp://osisaf.met.no/prod/ice/edge/
Archive HL FTP server : ftp://osisaf.met.no/archive/ice/edge/
EUMETCast EDC
Verification/validation method :Comparison with high resolution manual ice charts
Applications and users
Usage :NWP and Ocean/Ice models, operational Met and Sea Ice services.
History
Authorised to be operational / released :
Declared operational / released since :
On EUMETCast since :
On EDC since :
On FTP server since :
First date :
Comments
Using Metop-B and -C, instead of Metop-A and -B (in OSI-402-c)
Switching from NetCDF3 to NetCDF4 as product format
Introducing associated uncertainty fields. The uncertainty field replaces the previous ‘confidence level” variable which will disappear. The confidence level can be reconstructed from the new uncertainty variable.
Using Metop-B and -C, instead of Metop-A and -B (in OSI-402-c)
Switching from NetCDF3 to NetCDF4 as product format
Introducing associated uncertainty fields. The uncertainty field replaces the previous ‘confidence level” variable which will disappear. The confidence level can be reconstructed from the new uncertainty variable.