Development of a reliable wind database for eastern Canada.

Any inferences resulting from the analysis of meteorological records lean on the robustness of the observational data. Uncertainties in the data can be roughly classified into two groups: uncertainties due to random and systematic errors. Random errors are unavoidable and inherent to the very nature of the measurements as instrumental responses to various physical phenomena. Systematic errors, are produced by scale shifts of the instruments or by some more or less persistent factor that is not accounted for. Therefore, it is essential to develop procedures that allow to identify and correct the observations for these errors thereby improving the quality of the data sets.