A team of expert scientists, meteorologists, and mechanical engineers in the U.S. are exploring the possibility of using drones or, unmanned aerial vehicles (UAVs), to collect invaluable weather data by flying them near or, into severe developing storms, with the ultimate aim of getting longer tornado predictions, and earlier tornado warnings to families and their homes who may find themselves in the tornado's destructive path. Collecting data such as temperature, pressure, and wind velocity at various points along the storm would help meteorologist determine the likeliness of a storm developing into a tornado.
Data collection would occur, either by having drones with payload delivery mechanisms that would release dropsondes (parachuted sensors that can collect data about the vertical profile of a storm), or by flying swarms of data-collecting drones into various points of a developing storm. The biggest challenge with current weather balloons and sensor packages used to predict a tornado from inside a storm, and the biggest difference with drones, is the inability to steer them once dropped. On the other hand, drones would be able to fly through various points of the storm, to get a more accurate and complete picture of its development.
The main challenge currently confronting the team of experts is to effectively develop a lightweight drone prototype capable of withstanding the type of rain, hail, and 120 miles per hour winds typically encountered in violent tornado bearing storms. A kevlar-type composite material, similar to what is used in bullet-proof vests, could offer a solution. Other challenges currently facing the team of experts include: drone flight readiness (it currently takes up to four hours to prep their drone for takeoff); crashing computer systems; miscalibrated sensors; and drone flying restrictions that limit data collection.