Over the past few years Environment Canada, in collaboration with Hydro-Quebec, has developed, run and evaluated SPEO (Système de Prévision Eolienne), a High-resolution Wind Energy Forecasting System. SPEO makes use of GEM-LAM and it is configured to run over a 2.5km grid. In an extension of this project with the collaboration of the University of New Brunswick, the Wind Energy Strategic Network and the New Brunswick system operator, research is being done to investigate smaller domains, centered over individual wind farms. The primary goal of this project is to determine if these low cost and computationally efficient system can result in a reliable high-resolution wind power forecast.

Wind power forecasts rely primarily on wind, and to a lesser extent on temperature and humidity. Preliminary results from our experiments show that on average the micro grid can reproduce the wind forecasts produced using a larger domain without a significant change in the skill. With proper forcing of cloud information the surface temperature forecasts can also be reproduced with a micro grid. These results are quite promising. Further tests are required to evaluate the applicability of micro grids to other weather related needs and phenomena.