Models, Data, and Methodology - On the Effect of Offshore Wind Farms on the Atmosphere and Ocean Dynamics - page 21

a b c d

Fig. 3.3 Arrangement of wind turbines in TOS-01. Wind turbines are placed within grid boxes of
3km 3 km dimension and are marked by red diamonds . Box (a) shows the arrangement of a
small OWF consisting of 12 turbines like alpha ventus ,(b) consists of 48 turbines, (c) has
80 turbines, and (d) is composed of 160 turbines. The effect of each turbine counts ratable into
corresponding grid boxes. Spaces between wind towers count 500 m in x - and y -directions in (a)
and (b) and 1,500 m in x -direction and 1,000 m in y -direction in (c) and (d)

40 m up to 120 m. Although alpha ventus deals with bigger wind turbines, the
smaller type of turbine is chosen based on the statistics of BWE.

Model Setup in TOS-01 (Model Box)
The model setup of the model box differs between simulations of the atmosphere
and ocean. Thus, the model setup description is separated into METRAS and
HAMSOM. The distinction is a result of an indirect coupling between the atmo-
sphere and ocean by using METRAS data as meteorological forcing for HAMSOM
and the assumption of idealized model conditions.

METRAS (Atmosphere Box Simulations)
As mentioned, METRAS simulations were done in collaboration with the Meteo-
rological Institute of the University of Hamburg supported by M. Linde (Ph.D.
candidate at MI-Hamburg since 2011). The aim of the simulations done was the
creation of mainly a wind field including OWF information, which can be used as
forcing for HAMSOM. Therefore, the METRAS simulation differs from the
HAMSOM simulation, especially in time.

3 km. In the vertical,
METRAS has an equidistant resolution of 20 m below 100 m, and above 100 m the
vertical dimension of grid boxes is spread by a factor of 1.175. That vertical
resolution is used for all simulations. The top is placed in 9,521 m.
As mentioned, the model bottom is flat due to the use of an ocean landscape type
with a constant sea surface temperature (SST) of 15 C over the whole simulation
time. Additional variables prescribed for simulations are

METRAS is used with a horizontal resolution of 3 km

relative humidity of
70 % at 10 m, an one-dimensional start field that is chosen as stable dry, and
geostrophic wind field. The forcing of METRAS only affects directly the upper
layers above 1,000 m and only at the horizontal boundaries. Therefore, METRAS
simulates the boundary layer by itself.
The geostrophic wind ug was prescribed with a velocity of ug

a

' '

¼ 8.0 m/s,

respectively 5.0 and 16.0 m/s for analysis of the OWF s effect based on different

'