C5. Representation of large lakes in weather and climate models

Lakes are important components of the climate system and can affect regional climate by modulating surface albedo, surface energy and moisture budgets. The earlier versions of CRCM did not have lakes, except for the Great Lakes that were modelled using a mixed-layer lake model with thermodynamic ice treatment. Based on the offline analysis of available lake models (Martynov, 2010), it was decided to retain two lake models, the Hostetler model (Hostetler, 1993) and the Flake model (Mirinov, 2007), which were implemented in CRCM5. Results showed that both formulations are able to adequately simulate the lake surface temperatures in the case of shallow lakes. However, for deep and large lakes such as the Great Lakes, the model was not able to simulate observed temperatures, due to lack of representation of dynamical processes in their column framework. Three-dimensional dynamical models are required to realistically simulate the thermal regime for these lakes and therefore the fluxes at the lake-atmosphere interface. It is proposed here to implement NEMO (Madec, 2008) to model large lakes (such as the Laurentian Great Lakes and Great Lakes of the Mackenzie River basin) and their impact on regional climate. The choice of the model NEMO is based on the fact that the same model is used in the NWP GEM model at EC. This work will be done in collaboration with scientists at ‘Recherche en Prévision Numérique’ (RPN) and CCCma, since they have the experience of dynamical ocean coupling. The current horizontal resolution of CRCM5 (~10 km) is approaching that of 3D-ocean/lake models (2-5 km), making the coupling realizable.

The coupled model will be used to address science questions such as: How Great-Lakes modulate regional climate? How does the regional hydrology change from current to future climate? How well the high-resolution coupled simulations capture lake-effect snow?