We use statistical methods to assess the uncertainty of the DOWA climatological parameters, and compare it to the uncertainty in the KNW atlas. In particular, the following aspects are studied:
x The precision of the DOWA and KNW wind data. Precision of a data source refers to the random (non-systematic, non-predictable) error. Bias can be corrected for, but a lack of precision introduces an uncertainty that cannot be avoided. We use triple collocation and quadruple collocation analysis to assess the precision.
x The relative bias (compared to Meetmast IJmuiden mast and lidar measurements) of the DOWA and KNW atlas data. The bias in wind speed is assessed by means of quantile-quantile (QQ) plots that provide the relationship between the sorted values from one data source and the sorted values from another data source. Also, the bias in wind shear is assessed using QQ-plots.
x The uncertainty in wind resource assessments based on 10 years of wind data. To assess this uncertainty, we use a bootstrap method, both with and without accounting for multi-year dependence.
C. de Valk and I.L. Wijnant. Uncertainty analysis of climatological parameters of the Dutch Offshore Wind Atlas (DOWA)
KNMI number: TR-379, Year: 2019, Pages: 34