The test plan devised jointly by GMTB, EMC, the NGGPS Sea Ice Modeling workshop committee and interested workshop participants, included experiments over a one year period, with CICE being run in a standalone framework, forced by atmospheric and oceanic fields from the NCEP operational Climate Forecast System version 2 (CFSv2). GMTB conducted the model runs, and evaluation was performed jointly by GMTB, EMC, and ESRL. The experiment had three phases as described below, with varying model resolution and ways of initializing the atmosphere and ocean fields, as well as of constraining the Sea Surface Temperature (SST).

Phase 1: CICE resolution: 30 km at the pole, atmospheric initialization and forcing from CFSv2 1 deg analysis, ocean initialization from CFSv2 1 deg analysis, ocean forcing from CFSv2 1 deg analysis at 6-h intervals.

Phase 2: CICE resolution: 15 km at the pole, atmospheric initialization and forcing from CFSv2 0.2 deg analysis, ocean initialization from CFSv2 0.5 deg analysis, ocean forcing from CFSv2 0.5 deg analysis at 6-h intervals.

Phase 3: CICE resolution: 15 km at the pole, atmospheric initialization and forcing from CFSv2 0.2 deg analysis, ocean initialization from CFSv2 0.5 deg analysis, freely evolving ocean.