m 
Fe) 
T. Spangehl et al.: Intercomparing the quality of recent reanalyses for offshore wind farm planning 
FINO1 
annual mean wind speed [m s‘'] 
.( 
10 
7, 
A 
Dataset 
REA6 100M — 
REA2 100m == - 
CERRA-fe 100m +++" 
ERA5 100m —— 
HoKliSim-De 100m - - - 
NEWA 100M — 
FINO1 102 m = = 
9.5 
35 
a nn nn NND, DD, DT 
9.76 CN) 02 0 %0,%O, 0,50, %D, 
CA ACC ON CN N A 
time [vear] 
FINO1 
annual mean wind speed [m s‘!] 
%) 
m 
10.5 
s 
10 
Fa 
" 
- 
N 
Dataset 
REA6 100m — 
REA2 100m == = 
CERRA-fe 100m +++ 
ERA5 100m —— 
4oKliSim-De 100m - - - 
NEWA 100M — 
FINO1 102m == = 
a5 
Q 
35 
. 
Q < 
Yo Yo % 7 
time [year] 
Q, 
Figure 1. Annual mean wind speed at 100m for model based products at FINO1 grid points. Results are shown for (blue solid curve) 
COSMO-REAG, (blue dashed curve) COSMO-REAZ2, (purple dotted curve) CERRA-Tfc, (cyan solid curve) ERAS5, (cyan dashed curve) 
HoKliSim-De and (red solid curve) NEWA. Results obtained from FINO1 measurements at 102 m are additionally shown (bold black dashed 
curve). The time period is (a) 1979-2020 and (b) 2004-2009. 
Table 2. Bias of annual mean wind speed [m s” 11 at 100 m for model based products at FINO1 grid cells. FINO1 anemometer measurements 
at 102 m are used as reference. NA stands for not available 
Bias of annual mean wind speed [m s711 at 100m at FINO1 for different products 
Year 2004 2005 2006 2007 2008 2009 2004-2009 
COSMO-REA6 -—0.03 0.22 0.16 0.14 0.22 0.04 
COSMO-REA2 NA NA NA 0.35 0.44 0.18 
CERRA-fc —0.01 —0.07 0.00 —0.08 0.19 0.01 
ZRA5S —0.22 —0.32 —0.31 -—0.28 -—0.08 -—0.21 
AoKliSim-De —0.08 —0.07 —0.13 —0.13 —0.01 —0.17 
NEWA NA -—0.38 -—-0.29 —047 —0.19 —0.26 
NW, NNW), where each sector covers an angle of 22.5°, are 
used to analyse the frequency distribution of the wind direc- 
yon at heights near 100m. The analysis is performed for the 
time period 2004 to 2009. First, results for the whole year 
are discussed (Fig. 3). In a second step, the statistics of the 
Adv. Sci. Res., 20, 109-128, 202 
four seasons (DJF, MAM, JJA, SON) are briefly addressed 
(Fig. Al). 
The wind rose of the observations at FINO1 shows a max- 
imum relative frequency of nearly 10 % at the SW sector and 
slightly lower relative frequencies of about 9 % at the neigh- 
bouring sectors SSW and WSW (Fig. 3e). Lower relative fre- 
httos://doi.org/10.5194/asr-20-109-2023