Evaluation of Ice Models in Large Lakes

PI: Kevin Lamb, University of Waterloo

Co-I's: Marek Stastna; Andrea Scott; University of Waterloo

The primary goal of this project is to compare and validate the ability of two existing ice models to simulate the evolution of ice cover on large lakes at large and small scales. The nature of Ice cover in large lakes is very different from that in small lakes in that (i) large lakes are typically only partially covered; and (ii) ice in large lakes is often fragmented and drifts around the lake under the action of wind. Simulations using two ice models, both of which include snow, will be carried out using the same hydrodynamic core, so that differences observed can be attributed to differences between the ice models, as opposed to the manner in which the hydrodynamics is represented. We will use the MITgcm as the hydrodynamic core. The first ice-model will be the ice-model included in the MITgcm. The second model we will use is the Los Alamos Sea Ice Model (CICE). Both large scale simulations of entire lakes and small scale process studies will be undertaken. The small scale process studies will focus on lake ice dynamics and convection near the ice edge and under ice. We will also add a Lagrangian model to simulate three-dimensional frazil ice formation.

The primary location of application of the model evaluation will be Lake Erie. The partial ice cover common to Lake Erie winters, the differences between the three basins that make up the lake, and the rich biogeochemistry make this the ideal choice for this study. Secondary focus regions will be, Lake Ontario, and the outflow of the fast flowing, highly turbulent, Niagara River.