Teams of international researchers are converging on a valley in eastern Portugal this spring to capture granular data on wind moving over rough terrain. The goal is to develop high-quality reference data to validate computer models used in the siting, design and operation of wind turbines. Further, it is expected to aid in the prediction of how air pollution settles into valleys and in the navigation of aircraft over gust-prone, mountainous landscapes.

“We are going to cover with instruments a valley 7-kilometers-long by 2-kilometers-wide and, [for] the area outside, out to 1 kilometer,” says Harindra Joe Fernando, a fluid-dynamics engineer in the University of Notre Dame’s Dept. of Civil and Environmental Engineering and co-leader of the U.S. researchers. The U.S. team is funded by $3.4 million from the National Science Foundation. Teams from the eight European countries involved are funded by the EU. The project is described in the Feb. 16 issue of Nature magazine.

Existing wind models typically are validated by measurements from a 1980s field experiment conducted on and around a hill in Askervein, Scotland.

“The Askervein experiment is only for a very simple case,” says Fernando. “That’s not bad, but this one really has a huge deployment of high-end instrumentation.”

The setup for this experiment is named after a nearby town, Perdigão. The project includes 54 masts of 10-meter, 60-m or 100-m-tall instrument-laden towers arrayed along and perpendicular to the 200-ft-tall parallel ridges lying crossways to the prevailing winds. At 20 times per second, the devices measure wind direction, speed, temperature, humidity and other factors. Another 22 instruments will collect readings of small-scale wind flow in three dimensions, using the laser-based technique of light detection and ranging (LiDAR) and a sonic variant (SODAR), which is used as a wind profiler to measure the scattering of sound waves by atmospheric turbulence.

Fernando says the LiDAR units will sweep a hemisphere of sky out to 5 km, capturing at a 25-m resolution the 3D readings where the beams intersect. “It shows the instantaneous velocity and turbulence,” he says. “We cover the whole 180º in about three minutes. When you do that, you don’t have to make any assumptions.”

“The European idea is to get a good set of data that they can use to run models accurately,” Fernando says. “Their project is to map the winds for [all] Europe in 100-kilometer scale, and they have to use models with good fidelity. To do that, they need to validate the data.”

The U.S. researchers are focusing on turbulence, which can significantly rob wind turbines of generating efficiency and damage gearboxes. Portugal has a well-developed wind-energy infrastructure that includes one turbine conveniently sited in the test area, Fernando says.

“They found this beautiful valley that has parallel ridges,” Fernando says. “It’s quite ideal.”