T. Spangehl et al.: Intercomparing the quality of recent reanalyses for offshore wind farm planning 121 Figure 5. Bias distribution of near-surface wind speed [m s?1] for grid points embedded in geographical area covering the German Exclusive Economic Zone of the North Sea. The spatiotemporal bias distribution is illustrated by (top row panels) Box–Whisker plots for the two different satellite based reference data sets (left panel) Scatterometer and Model (e5) from CMEMS and (right panel) HOAPS from CMSAF. Values outside the whiskers (whisker length never exeeds 1.5 IQR (interquartile range)) not shown. All calculations are based on monthly data. Geographical area is 3–8.5? E, 53.5–56? N. Maps of the multi-year averaged bias of (from 2nd to 5th row panels) ERA5, COSMO- REA6, CERRA and HoKliSim-De compared to (left panels) Scatterometer and model (e5) from CMEMS and (right panels) HOAPS from CMSAF for the considered geographical area are additionally shown. Black contours display the geographical position of the German Exclusive Economic Zone. FINO1 is additionally shown. The time period is 2008–2017. https://doi.org/10.5194/asr-20-109-2023 Adv. Sci. Res., 20, 109–128, 2023