Public Release: 

Optical Sensor Measures Wind Direction Over Long Distances

Georgia Institute of Technology

A prototype, non-Doppler optical sensor that makes inexpensive, accurate measurements of crosswind speeds over long distances holds promise for chemical manufacturing, aviation safety and meteorology.

The single-ended, long-path laser wind sensor registers faint wind movements that an anemometer cannot measure. Test results show its measurements of higher wind speeds correlate with those of anemometers, said Dr. Mikhail Belen'kii of the Georgia Tech Research Institute (GTRI).

"The sensor is more sensitive and accurate than mechanical anemometers, and it may provide an advantage when monitoring winds over a wide area, by providing a low-cost alternative to complex arrays of traditional sensors," said Belen'kii, a principal research scientist in the Electro-Optics, Environment and Materials Laboratory. "With some modification the sensor can measure both horizontal and vertical wind speeds."

The sensor, originally developed for chemical plants, is designed to work alongside other sensors that measure airborne chemical concentrations, said principal research scientist Dr. Gary Gimmestad.

"If you measure the concentration and the crosswind at the same time, you can get a good idea of the rate at which a pollutant is leaving a plant," he explained.

But because the sensor measures average wind directions over long distances, it might have additional applications in aviation, meteorology, or aerosol dispersion studies. It would be particularly useful in locations where erratic winds are the norm -- tank farms, cities or widely varying landscapes. A provisional patent application has been filed on this work.

The sensor's design is simple. All optics and electronics are mounted on a large telescope. An inexpensive helium neon laser about two inches in diameter projects a beam of light from this unit onto a target approximately 100 feet away. The target is made of retroreflective materials used on highway signs.

The method the sensor uses is based on a laser beam degradation phenomenon known as the residual turbulent scintillation effect. The telescope collects laser light reflected by the target, and sends it through the series of optics. Among those optics are two tiny, horizontally separated detectors, each of which monitors a spot on the target inside the laser beam. The detectors pick up shadowy waves, or fringes, moving across the laser beam. The waves are visible on the target material, said research scientist David Roberts.

"The fringes look a lot like the shadows of waves created on the bottom of a swimming pool on a sunny day," Roberts explained. "If you look at turbulent wind using a laser beam, you see something very similar to those waves traveling across the beam."

Each of the two detectors in the sensor registers the moment at which a dark fringe passes its view. By digitizing the points at which each detector picks up a single wave, a computer can measure time and separation. It then can compute the average velocity of a massive column of air crossing the laser beam. In this case, wind speed calculations were made every 10 seconds.

"Even though air may be flowing erratically -- some going one direction at one end of the beam and some going exactly the opposite direction -- you can get a net flow across the laser beam with this method," Roberts said.

The sensor correlated extremely well with anemometer readings in test results with 100 feet between the sensor and the target. A five percent discrepancy, within the limits of experimental error, was observed between wind speed measurements by the sensor and by the anemometer in laboratory tests.

Companies often rely on anemometers to check wind direction and velocity. But the anemometer measures wind in just one location -- and in situations where winds blow erratically, that may not be representative of overall wind movement in a larger area. To truly duplicate the work the prototype sensor performs, a row of anemometers would have to be placed side by side in a line as long as the laser, Belen'kii said -- a very expensive proposition.

"If you collected the same information using several meteorological towers, it would cost you much more," he said. "The cost of this sensor would be less than that of one meteorological tower."

The sensor is easier to use than Doppler systems, the researchers say. In addition, it measures wind across the beam of light instead of along the beam, as Doppler systems do. And unlike conventional LIDAR systems, this sensor can pick up turbulence.

"This might prove to be a better and more accurate way of measuring turbulence," Belen'kii said.

"It's really pretty robust, too," Gimmestad added. "It operates well in sunlight and in darkness."

Like many optical systems, the sensor doesn't work as well in rain or fog, which obscure its target. And the sensor can only measure the component of wind that crosses the laser beam at right angles. However, one sensor could be made to rotate among several targets, checking air movements at a variety of angles.

Researchers next plan to test the sensor with technologies that measure airborne pollutant concentrations at a real refinery plant. Additional testing might include tracking crosswind speeds from the tops of city buildings, and modifying the sensor slightly to measure cross winds and wake vortices along runways at airports. The sensor also could be configured to measure vertical winds, which would provide an unusual, three-dimensional capability.

Varying the laser wavelength, power and beam geometry, target type and range, receiver diameter and data processing algorithms could make the sensor useful in additional areas, as well.

A poster paper on this work was presented at the International Symposium on Optical Science, Engineering and Instrumentation in Denver during August. The work was funded under the GTRI Internal Research Program.


430 Tenth St. N.W., Suite N-112
Georgia Institute of Technology
Atlanta, Georgia 30318

John Toon (404-894-6986);
FAX: (404-894-6983)

Dr. Mikhail Belen'kii (404-894-0140), Internet:;
Dr. Gary Gimmestad (404-894-3419), Internet:;
David Roberts (404-894-3493), Internet:

WRITER: Lea McLees


Disclaimer: AAAS and EurekAlert! are not responsible for the accuracy of news releases posted to EurekAlert! by contributing institutions or for the use of any information through the EurekAlert system.