Algorithm Input and Output Data Definition (IODD)

Input Data

Input L1 data

Field	Description	Shape/Amount
L1B TB	L1B Brightness Temperature at C, K and KA-bands (both H and V polarization)	full swath or section of it (Nscans, Npos)
L1B TB OZA adjustment	L1B Brightness Temperature at C, K and KA-bands (both H and V polarization)	full swath or section of it (Nscans, Npos)
L1B NeΔT	Random radiometric uncertainty of the channels	full swath or section of it (Nscans, Npos)
L1B Geolocation	Latitude / Longitude	full swath or section of it (Nscans, Npos)
L1B OZA	Observation Zenith Angle	full swath or section of it (Nscans, Npos)

Auxiliary data

Field	Description	Shape/Amount
land mask	A land mask	Remapped to dimensions of the swaths (can also be available in Level-1b files).
maximum ice	Climatological maximum ice coverage	Remapped to dimensions of the swaths.
tie-points	Algorithm Tie-points (algorithm coefficients) are generated offline (e.g. once a day) as part of the Level-2 SIC chain.	a few text files per day

Additional auxiliary data will be required if RTM correction of the brightness temperature is activated. In that case, fields of T2m, T0m, Wind Speed, Total Column Water Vapor, Total Cloud Liquid Water, remapped to the dimensions of the swaths and collocated in time, will be required. They can for example come from ECMWF operational analysis and forecast system.

Input L2 data

The SIC and SIED algorithm does not require fields prepared by other L2 processors as input.

Output Data

Field	Description	Shape/Amount
sea-ice concentration	The SIC between 0 and 100% (0%: Open Water, 1%: Consolidated Ice)	EASE2 grid (nx,ny)
sea-ice edge	The SIED binary mask: 0 or 1 (0: Open Water, 1: Sea Ice)	EASE2 grid (nx,ny)
raw sea-ice concentration value	SIC before Open Water Filter and thresholding (can be < 0% and > 100%)	EASE2 grid (nx,ny)
sea-ice concentration total uncertainty	The retrieval uncertainty as 1-sigma	EASE2 grid (nx,ny)
sea-ice concentration uncertainty components	Several components of the retrieval uncertainty (TBD, e.g. radiometric uncertainty and algorithm uncertainty)	EASE2 grid (nx,ny)
status flags	indicates the reasons for missing, bad, or nominal SIC values.	EASE2 grid (nx,ny)

Important

In the table above, we used ESA2 grid (nx,ny) for the shape of the output fields, thus a gridded SIC and SIED L2 product. This corresponds to the current implementation in the later sections of this ATBD. However, we note that most of the SIC (and SIED) algorithm is applied in swath projection. Later versions of the algorithm could implement a swath projection as it output Shape/Amount.