Could crop roughness impact the wind resource at agriculturally productive wind farm sites? B.J. Vanderwende1 J.K. Lundquist1,2 1. Atmospheric and Oceanic Sciences University of Colorado Boulder, CO USA 2. National Renewable Energy Laboratory Golden, CO USA We would also like to acknowledge collaborators at the Los Alamos National Laboratory for the computational time required to run the following simulations

3rd ICEM Meeting Tuesday, 22 June 2015

Can crop selection impact the power available to wind turbines aloft?

2007 Census of Agriculture, U.S.D.A.

Q1 2014 U.S. Wind Power Capacity, AWEA

Maize/corn

Soybeans

2

Motivation Iowa is a major center for both agriculture and wind power in the United States

Hypothesis

Switching from corn to soybeans can increase power output; even in a

wind farm

3

Fully grown corn/maize

• Heights of 2 - 3 m

• Davenport classification* roughness value of 25 cm

Fully grown soybeans

• Heights of 1 - 1.5 m

• Davenport classification* roughness value of 10 cm

*using modified table from Davenport et al. 2000

4

Model Methods

Crop roughness used for mature growth stage, common in late summer months

5

Model Methods

We modify the land representation in WRF model to study crop impact on winds

Common WRF settings

• Version 3.4.1 ARW

• Nested 1.25km domain

• 60 vertical levels

• Noah land surface model

• ERA-Interim input data

• MYNN 2.5 PBL Scheme

Wind farm

Corn crop

Free stream

Soy crop

Free stream

Corn crop

Wind farm

Soy crop Four model runs spanning

full month of August 2013

• Hypothetical 121 Vestas V90 turbine wind farm

– 1.8 MW, 80 m hub height, 90 m rotors

– Minimum 7D spacing used in square grid

Full details on WRF-WFP in Fitch et al. 2012 6

Model Methods

Wind Farm Parameterization used to estimate crop impact in presence of wind farm

Elevated momentum sink removes energy independently of surface state in the model

Roughness approach

Elevated sink approach

• Mason (1988) found the height of a z0 influence scaled approximately as: – H ~ 200Lc

– Lc = distance from roughness boundary

• Setting H to top of turbine rotor (125 m), Lc should be 25 km

7

Model Methods

Size of roughness patch determined using boundary layer growth theory

Modified z0 patch area 56.32 km2

Model wind farm area 6.32 km2

Mean wind direction

8

Wind Impacts

Roughness effect on wind speed and turbulence decreases rapidly with height

Soybean – corn

Rotor layer

9

Wind Impacts

Roughness effect on wind speed and turbulence decreases rapidly with height

Note that TKE is not actually represented by the mesoscale model grid

Soybean – corn

Rotor layer

10

Wind Impacts

Stability strongly influences vertical mixing in the boundary layer

Soybean – corn Free stream – wind farm

Increasing stability

Rotor layer

11

Wind Impacts

Stability bins can be defined to easily summarize trends across the entire wind farm

Stability Class BRN Range

Convective RB < -0.05

Near-neutral -0.05 ≤ RB < 0.05

Stable 0.05 ≤ RB < 0.3

Very stable 0.3 ≤ RB

Note that roughness effect is not significant in the most stable conditions

12

Wind Impacts

Large percentage changes in hub-height wind speed and rotor-layer shear across the farm

Shear represented using power law

coefficient

Stability Class BRN Range

Convective RB < -0.05

Near-neutral -0.05 ≤ RB < 0.05

Stable 0.05 ≤ RB < 0.3

Very stable 0.3 ≤ RB

𝑢1𝑢0

=𝑧1𝑧0

𝛼

𝑧0 = 35𝑚, 𝑧1 = 125𝑚

Soybean – corn

13

Wind Impacts

Transects aligned with the wind direction show spatial impact of stability on IBL growth

Only takes 10-15 km for effect to build to top of rotor layer

Can more practical patch sizes be used?

Wind Direction

Soybean – corn

14

Wind Impacts

One-week sensitivity simulations suggest smaller patch sizes could be viable

202 km2 patch produces 19% weaker impact on hub-height winds than original size

Realistic, or a result of WRF diffusivity?

15

Potential Economic Benefits

Simulated winds and turbine power curve were used to estimate effect on farm output

• Use of soy instead of corn yields 14% increase in energy production over month

– From 3% UH change

• Monetary benefit given $30-60* per MWh

– $133-266k for August

* Wiser & Bollinger 2012

Soybean - corn

Summary: crop selection can impact wind farm output

during peak of growing season

• For August 2013, ~14% difference in power output for wind farms over soybean vs corn crop

• Corn at full height for 50-60 days and soybeans for up to 90 days

– Annual effect depends on full cycles

• Agricultural land use decisions subject to many considerations

16

Thank you for your attention!

Any questions?