Using small drones to measure wind speeds in the polar regions

Drones and similar small unmanned aerial vehicles (sUAVs) have seen a massive surge in popularity over the past few years due to their innovative applications, such as crop monitoring, search and rescue operations, and coast profiling. The potential of sUAVs in atmospheric science and meteorology has not gone unnoticed either as drones offer an efficient way to place various kinds of sensors up above in the lower atmosphere.

As evidenced by pioneering studies led by the National Institute of Polar Research, Japan, small drones bearing meteorology sensors can be useful for weather forecasting in places where weather stations are scarce or costly to set up and maintain, such as the polar regions. With the use of drones, invaluable data could be gathered for making numerical weather predictions and improving climate models.

However, several practical challenges remain to be addressed on this front. One clear example is measurement of wind speeds for wind profiling. While a few companies are commercializing drones for conducting wind speed measurements, these drones tend to be large and expensive. This is because the ultrasonic anemometers that they carry are heavy and bulky. On the other hand, although wind speed could in theory be calculated from the drone’s internal flight log data, there is no way to access these data in most commercial drones. Thus, with no alternative methods to measure wind speed, application of sUAVs remain limited.

To tackle this issue, Associate Professor Jun Inoue of the National Institute of Polar Research and Assistant Professor Kazutoshi Sato of the Kitami Institute of Technology, both in Japan, have recently conducted a study in which they tested and validated a new wind speed measurement device compatible with sUAVs: thermal anemometers.

As explained in their paper published in MDPI’s Drones published on October 3, 2022, thermal anemometers, also called “hot-wire anemometers,” are small electronic devices that can indirectly measure wind speed based on how fast a heated wire cools down over time due to air flow. These sensors are ideal for sUAVs as they are extremely lightweight (~ 1 g) and inexpensive (

The researchers tested their approach in the lab as well as in the field. In the laboratory experiments, they flew a small quadcopter drone equipped with a thermal anemometer in front of large fans. Using smoke machines, high-speed cameras, lasers, and image analysis software, they analyzed how the air flow was disturbed by the drone as it ascended. Using these data, they created a bias correction algorithm, which enabled the drone to make more accurate wind speed measurements.

For the field experiments, the team flew their drone over the ice-covered Okhotsk Sea while aboard the Japanese Patrol Vessel Soya to make wind speed measurements. Upon comparing the data gathered by their drone with that from an ultrasonic anemometer on the ship, the team found them to be in agreement, suggesting that the drone’s measurements were highly reliable. “The root-mean-square error was ±1.13 m/s, suggesting that the proposed correction method was accurate and well within the requirement for numerical weather predictions,” remarks Dr. Inoue.

Overall, this alternative approach to wind profiling is definitely promising on various fronts. Additionally, the researchers also proposed two more techniques that could be pursued in future works: calculating both wind direction and speed using two thermal anemometers and transferring the gathered data in real time.

Wind speed measurements are not just relevant for climate science and weather forecasting. The drones themselves can benefit from knowing the wind conditions they are flying in as it enables them to stabilize their flight better. “The flight style of drones has changed from flying within the operator’s visual line of sight to flying beyond the visual line of sight. This is because they are now routinely used for long-range delivery and surveys of disasters and forest fires,” says Dr. Sato.

“Once we can monitor the wind conditions in all kinds of drones in real time, operators will be able to focus more on their task without having to worry about the drone’s stability and safety,” he concludes.

Provided by
Research Organization of Information and Systems