A Mavic Pro drone is flown on Jan. 18 in Bridgend, Wales. The U.K. government has moved to give police further authority to tackle illegal use of drones, including powers to land, seize and search drones, following a drone incursion at Gatwick airport in December. (Photo by Matthew Horwood/Getty Images)Getty

Gatwick, Heathrow, Newark, Dubai and today, Dublin: Each of these airports has been in the news recently when flights were halted or delayed by sightings of what were believed to be drones in the area.

So how big a threat is this? Are drones a danger to manned aircraft?

With 1.3 million drones now registered with the Federal Aviation Administration (FAA), up from about 470,000 in 2016 when drone registration was first required, anyone can see that there are more drones in the air than ever before. While a small percentage of these drones are operated for commercial purposes by FAA-certified remote pilots, the vast majority are operated by hobbyists for fun and recreation. Hobbyist pilots are required to fly under the safety guidance of a model aircraft organization, like the Academy of Model Aeronautics (AMA), and they have to keep their drones in sight, below 400 feet, and out of airspace meant for passenger-carrying airplanes. Commercial drone pilots have to know and abide by similar rules, but unlike hobbyists, they have to take an FAA test to prove it.

Because there is currently no test for recreational drone pilots, the FAA, the AMA, and a regular alphabet soup of companies and organizations have tried to make sure that drone operators know the rules before they fly. DJI, the market leader in consumer drones, has included a knowledge quiz that pilots must take before they can unlock their new drones and fly. Predictably, the answers are on YouTube, but operators still have to read and answer the questions. But as anyone who has slowed down when seeing a police car knows, knowing the rules doesn’t mean always obeying them. Drone manufacturers are aware of this, and try to use software solutions to keep drones away from the areas they don’t belong. DJI and other companies uses ‘geofences’, which alert pilots if their drones are in areas that are off-limits. In some cases, the geofences prevent drones from flying at all. AirMap is testing a new geofence system that will provide pilots with real-time audio and visual alerts if they are closing in on airspace that is geofenced. But geofences don’t always correspond to the airspace the FAA wants to protect, and users can often override or disable them.

So despite education and technological solutions, drones are sometimes found in places they do not belong. A recent study by Embry-Riddle Aeronautical University looked at drone flights over a two week period around Daytona Beach International Airport.  The results showed that 7% of drone flights tracked exceeded 400 feet, and 21% exceeded the recommended maximum altitude for the area in which they were operating. In one case, a drone was detected at an altitude of 90 feet within a quarter mile of the approach path to an active runway. In total, 8 drones were detected within one mile of the center of the airport.

For perspective, in that same time period, there were about 11,500 aircraft takeoffs or landings at that airport. By comparison, 8 drones isn’t much. On the other hand, that’s a lot of airplanes in the sky and a lot of potential conflicts.

How big people think the risk is depends on who you ask. Many drone operators have expressed skepticism that the Newark sightings were actually drones at all. Several cases have occurred in which objects reported as drones turned out not to be drones after all. Drone operators and airport personnel alike agree that drone sighting reports can be suspect. To be sure, drones are difficult to identify from the cockpit of an aircraft. In fact, they are difficult to detect at all. One study from Oklahoma State University found that even when they were looking for drones, the pilots of small aircraft detected drones only when they were a tenth of a mile away on average. But this might actually be cause for greater concern. Maybe it’s what pilots are not seeing that should worry us.

To find other manned aircraft, pilots no longer have to rely on just their eyeballs or air traffic control.  There are automated systems that help airplanes communicate with each other and alert pilots about potential conflicts.  Could something similar work for drones?  Yes and no.  Eventually, some form of identification and tracking will be required, but it will have to be a different system from what aircraft use now, or we risk overwhelming pilots and air traffic controllers with information and clutter.

If a drone was in the wrong place, your pilots didn’t see it, and your aircraft hit it, what would happen?  That question is not as easy to answer as you might think. Damage from a collision depends upon the mass of the object being hit and the impact velocity, but also the density of the object, the angle of impact, and the frangibility of the object, or how easily it breaks up.  That’s a lot of variables, but clearly the greatest damage would occur from a 90 degree impact of a heavy drone when the aircraft is moving at high speed.

Commercial airplanes are already required to withstand impacts from another airborne hazard – birds.  Specifically, transport aircraft have to be able to withstand impacts with birds weighing 4-8 pounds on the airframe at cruising speed, and ingested into the engine at takeoff power.  Admittedly, 4-8 pounds isn’t much, when you can legally fly a drone up to 55 pounds without any special permissions (as long as you abide by altitude and airspace restrictions). The good news is that the vast majority of consumer drones weigh much less than that, with the most popular models weighing in below ten pounds. But a series of studies performed at Virginia Tech suggest that the concentrated mass of drones compared to birds makes impact with a drone more damaging than that with a bird of a similar weight.  Given the unknowns, the best option is clearly not to hit a drone at all, but as we’ve seen, that’s not so easy to guarantee.

The ‘big sky, little airplane’ theory of keeping drones and airplanes apart breaks down when you put more and more aircraft in the sky, no matter how little they are. Whether you are a drone enthusiast who is tired of being blamed for everything, or a private citizen who is wary of the noise, nuisance, and potential privacy threats from drones, we all can agree that this technology has huge potential to improve our lives in a thousand different ways we can hardly yet imagine. Missing those opportunities because we let something tragic happen would be a loss for all.

And whatever our opinions are about drones, the truth is that in aviation, we don’t make decisions based on opinion. We make them based on facts. This data-driven, risk-based approach has created the safest transportation system the world has ever known. So what do we need? More data. We need to know where drones are operating, how they are being used, where the threats are, and how bad it is when you hit them. Only with more information can we really understand the threat from drones accidentally encountering aircraft.

As for drones that purposely want to interfere with aircraft, well, that’s another story.