A few of the Challenges Faced with the Upcoming Implementation of a 200-m Pan-Canadian External Land Surface Modeling System

Developed in the last 6 years, Environment Canada's new external land surface modeling system is currently being tested for experimental implementation at the Meteorological Service of Canada (MSC). Integrated at 100-m or 200-m grid spacing and using state of the art methods to specify surface characteristics, this system called GEM-Surf has been shown to produce more realistic surface and near-surface forecasts compared with MSC's current operational systems. The gains are particularly impressive over mountain regions and over cities, as these types of land surfaces most benefit from the increase in horizontal resolution and from better specification of surface characteristics, e.g., orography, presence of road and buildings.

The presentation will focus on a few remaining challenges for this type of system. One of them is the requirement for more sophisticated land surface schemes over natural surfaces (bare soils and vegetation), cities, and glaciers. Some details will be provided on the new land surface scheme called the Soil and Vegetation Simulator (SVS) currently being developed and tested in GEM-Surf, with a focus on aspects related to hydrology, as well as water balance and exchanges. Another challenge is related to the specification of initial conditions for GEM-Surf, which are obtained from continuous cycling forced with best estimates of atmospheric forcing (precipitation, air temperature and humidity, winds, downwelling radiation). Another aspect, often neglected, is related to the several methods still used in the community to specify momentum roughness in sub-km grid spacing models. The scale at which the effect of ‘unresolved' orography has to be considered is unclear, as a large portion of the turbulent exchanges between the surface and the atmosphere occurs at scales larger than the model resolution. These items will be discussed in the context of test runs over Southern Ontario, with a focus on summertime periods and with evaluation against surface observations as well as space-based remote sensing data.