Wind and turbulence at a forest edge

1
Wind and turbulence at a forest edge
Ebba Dellwik, Ferhat Bingöl, Jakob Mann and
Andrey Sogachev Wind Energy Division Risø
National Laboratory for Sustainable EnergyDanish
Technical University
2
Outline

• Introduction to scientific goals and experiment.
• Mast and lidar data analysis at different
  atmospheric stability conditions.
• Modelling / model introduction.
• Preliminary WAsP Engineering solution.

3
How can the forest be parameterised in simple
models?
Displacement height gt Forest edge effects
Problem areas
displacement height
4
Experiment
LIDAR 1
LIDAR 2
Masts for turbulence and mean wind speed
measurements
5

Site Tromnæs forest
FOREST
FIELD
6
(No Transcript)
7
(No Transcript)
8
(No Transcript)
9
Mean wind speed neutral stratification, summer
10
Mean wind speed neutral stratification, winter
Mean wind speed neutral, winter
11
Mast data stable stratification, summer
12
Turbulent kinetic energy neutral, summer
13
Turbulent kinetic energy neutral, winter
14
Experiment
LIDAR 1
LIDAR 2
Masts for turbulence and mean wind speed
measurements
15
LIDAR horisontal scan _at_ 30m
Relative wind speed (-)
16
SCADIS

• Solves the RANS
• equations
• k-eps closure
• Enables detailed forest
• parameterisation
• Originally developped
• for environmental
• applications

17
Forest parameterisation
18
SCADIS forest parameterisation
Forest type
Drag force S
Forest density
19
Mean wind speed neutral, summer
20
Turbulent Kinetic Energy neutral, summer
21
Model-LIDAR comparison
Relative wind speed (-)
22
2D modelling results
23
How can the forest be parameterised in simple
models?
Displacement height gt Forest edge effects
Problem areas
displacement height
18
24
WAsP Engineering
18 slope
25
Conclusions

• The relative speed-up over the forest edge is
  dependent on the atmospheric stratification as
  well as the canopy density.
• At both mast positions, the wind field is
  affected by the forest edge.
• The SCADIS model can predict the flow speed-up
  over the edge.
• WAsP Engineering comparison shows that
  substituting the edge with a slope of 18 degrees
  is a good first approximtion.
• Further analysis necessary to finalize
  recommendations for WAsP and WAsP Engineering.