Enhancing Great Lakes Ice Forecasting: Integration of CICE6 and the Unified Forecast System

The short-term prediction capability of coastal ice is critical for safe and efficient navigation, coastal facility management, and planning for winter recreations. In the North American Great Lakes (hereafter referred to as the Great Lakes), NOAA’s Great Lakes Operational Forecast System (GLOFS) provides nowcast and 5-day forecast guidance for 3-dimensional hydrodynamic conditions such as lake current, water temperature, and water level. The ice model, based on the unstructured grid version of Los Alamos Sea Ice Model (CICE), is added to the experimental version of GLOFS to provide 3-day forecast guidance for ice conditions (e.g., ice concentration and thickness). This ice model was cariblated and extensively verified with available observations, and is based on an older version of CICE (version 3-4). The latest version of CICE (version 6) has improvements in physics parameterizations and is fully compliant with the Unified Forecast System (UFS) infrastructure based on the Earth System Modeling Framework (ESMF) and the National Unified Operational Prediction Capability (NUOPC) layer. This work aims to demonstrate the capability of the latest version of CICE (CICE6) and the UFS infrastructure for applications to coastal ice predictions through testing for Great Lakes short-term ice predictions. The effort aims to enable future streamlined development of model components and flexible coupling with other model components (hydrodynamics, wave, and atmospheric data) using the UFS infrastructure.

A standalone CICE6 is configured for the Great Lakes at 1-km resolution, and driven by surface hydrodynamic data from GLOFS and atmospheric forcing from the High Resolution Rapid Refresh (HRRR). Results are compared with ice simulations from the experimental GLOFS for benchmarking purposes. In addition, sensitivity experiments are conducted with snow density and heat conductivity and ice thickness category definitions. Tests will be also conducted using the Community Data Models for Earth Predictive Systems (CDEPS) to force the standalone CICE6 model using the atmospheric data component.