Retrieving and comparing different sea ice forecasts (from operational forecasts and climate models) used to be a cumbersome task, but ECMWF has finally made our life simpler by developing a software package coined ICECAP (sea-Ice Calibration, vErifiCAtion and Products).
ICECAP is specifically designed to provide end-user forecast products based on sea ice forecasts from days to seasons ahead from multiple forecast centres, alongside information on the quality and uncertainty of those forecasts.
ICECAP can be used to retrieve and pre-process sea-ice cover fields from ensemble forecasts from C3S seasonal forecast models, TOPAZ5 forecasts (available through the MetNorway THREDDS server) as well as the corresponding sea-ice cover observations from C3S and CMEMS. More data sources will be added when they become available. ICECAP computes and visualizes a set of validation metrics and can also compute calibration statistics if necessary, i.e. for seasonal forecasts. This functionality allows scientific users to perform research on the forecasts and develop the forecasting systems.
ICECAP can also retrieve and pre-process sea-ice cover fields from a single (“real-time”) ensemble forecast, calibrate for known errors using calibration statistics computed in the steps above, and compute and visualize a set of user-relevant information products together with the information on forecast quality. This functionality is targeted at end users who need reliable sea-ice forecast information for planning their activities in potentially ice-infested waters.
The software therefore has two interfaces to cater for these two main user types. The scientific-user interface can run ICECAP on a Linux system using a workflow manager (e.g. ecFlow). This enables the user to use all implemented capabilities, i.e. verification, calibration, products. In addition, the end-user interface can be used via jupyter notebooks, however providing simpler capabilities.
The software can be found at https://github.com/ecmwf/ICECAP
ICECAP can be used to verify sea ice model results against observations. The Python package will be implemented on the WEkEO/EDITO platform to allow users to use ICECAP to easily obtain user-relevant products based on sea ice forecasts from
TOPAZ5 and selected C3S models.