In the present work, a mesoscale atmospheric model was used for short-term wind forecast for applications on wind energy generation. Different spatial resolutions and parameterizations were tested, and the comparison among the simulations was based on a methodology, which uses a set of statistical indices and error indicators widely accepted in the Meteorology community when evaluating atmospheric models. Such methodology was utilized in order to choose the best options for horizontal grid spacing, turbulence parameterizations and nudging timescale when representing the wind field over the coast of Ceará State, in Northeast Brazil, in four days during the transition period between the dry and wet seasons. The tests reveal that, among the different options, the model setup, which reproduces better the observations along the coast is the one with a 1km grid spacing, using an anisotropic Smagorinsky-type turbulence scheme and Newtonian relation timescale of 12 h.
wind forecast; atmospheric modeling; wind energy; statistical intercomparison
