Drone Sightings: The Actual Non-Hyped Numbers Analyzed

Jonathan Rupprecht

For years and years, many have been talking about the FAA’s drone sightings reports. The drone sightings reports have been cited many times by the news media and elected officials who were rather alarmed about the data.

Others in the industry cite the drone sightings as evidence of the greater need for the government(s) to do something by creating regulations. Some counter-drone companies have used it to show a need for their product.

Others say it is hype and the sightings data is greatly flawed and overblown because after all, they are only sightings, not impacts.

Regardless of where you come from in the industry and your motives, we need to accurately understand the drone sightings data. By the end of this article, you’ll understand the growth of drone sightings, where they typically are located, and whether the data is alarming as many have said.

Quick Summary of the Drone Sightings:

The reported drone sightings over time are NOT growing. They’re decreasing.
The FAA has inaccurately reported on the drone sightings data and this is proven by their own data they released(more on that later).
There are more drone sightings reported in populated areas than unpopulated areas.
There are more drone sightings reported in warmer months than colder months.
States with larger populations have more reported drone sightings.
There are more medium or large animal impacts with manned aircraft than mere reported drone sightings.
Any discussions we have on this topic should be using numbers and not just percentages or words.

Basically, population and weather/climate are the determining factors of when and where you’ll have drone sightings. The data also shows that we are past a peak in sightings and they are currently consistently decreasing.

But how many of these sightings are verified sightings of drones and not white balloons, seagulls, etc.?

How many of these reported sightings are of drones actually flying unlawfully, dangerously, or nefariously? Is there like a giant :( face on the drone that tells you it is bad?

A drone being flown by a good guy and one being flown by a bad guy are extremely difficult to tell apart.

This is where things get very interesting.

What Does the Word “Sighting” Mean?

Sightings are just that….mere sightings.

A pilot, person on the ground, law enforcement officer, etc. sees a drone and reports it to the FAA. It gets logged.

The drone could be lawfully flying, safely flying, carelessly flying, or flying with criminal intent. It’s all lumped into one thing – a sighting.

The accuracy of the reports cannot be verified. You could report you saw a drone all you want. No one can check. It’s literally a giant hearsay list.

And here is a very thought provoking question….is there anyway we can prevent people who would stand to benefit from higher drone sighting numbers from calling in false sightings?

I remember in lawschool my Criminal Procedure professor discussing anonymous tips and why they are by default very weak evidence. Basically, if anoynmous tips were given the same weight as reports from identified people to get warants or conduct arrests, then a law enforcement officer can easily just call in anonymously the data they need to do whatever they want. Ring ring….”911. What is the nature of your emergency?”……. “Uh….I’d like to report a suspsicious person who appears to be selling drugs.”

Likewise, who is to stop the FAA employees, FAA contractors, people working in counter UAS industry, or manned pilots from calling in more sightings that cannot be verified? Who is to stop over reporting where anything just gets reported as a drone?

Keep in mind that 18 U.S.C. 1001 makes it a criminal offense to make false statements to the FAA so many will be detered from doing that. Some of the reports have identifable information so they are not really anonymous and less likely to false report. On the other hand, the Department of Transportation’s Inspector General’s Office does investigate FAA officials for illegal activity and police arrest people all the time for fraudulent activity.

Provide what weight you will to the sightings. We are in a predicament. There is no evidence to prove that any of those groups reported falsely, but there is also no evidence to prove that everyone reported truthfully AND accurately. We need to be balanced in approaching this data.

Fact Checking the FAA’s Drone Sightings Data:

The Government Accountability Office’s report in May 2018 accurately summed the situation up, “FAA told us that most of the reports cannot be verified because a small UAS typically is not detected by radar, the small UAS pilot is usually not identified, or the small UAS or other physical evidence is not recovered. FAA and some aviation industry stakeholders also told us that the reliability of many of the reports is questionable; FAA explained that this is because pilots can have difficulty positively identifying objects as small UAS, given their small size, their distance from the observed position, the speeds at which a manned aircraft and a UAS are operating, or the various factors competing for the pilot’s attention. ”

UAS Weekly reported, “Unmanned Aircraft Safety Team (UAST) Drone Sightings Working Group released a new report on the Federal Aviation Administration’s (FAA) 3,714 drone sightings reports collected by flight crews, air traffic controllers and citizens from November 2015 to March 2017. The report found that only a small percentage of drone reports pose a safety risk, while the vast majority are simply sightings.” Report is located here.

The Academy of Model Aeronautics, as well as others, have brought up the drone sightings data. The AMA has multiple reports analyzing the data which calls into questions much of the data. Their analysis stated that some of the drone sightings are law enforcement flying the drones, balloons, or just harmless drone sightings.

Here is a striking quote from the AMA’s 5/10/2017 report, “In AMA’s previous analysis, AMA found that 3.5% (August 2015) and 3.3% (March 2016) of sightings included the explicit notation of’ ‘NMAC’ (near mid-air collision) or ‘near miss.’ Our analysis of the newest 1,270 records released in February 2017 show about the same percentage of near misses in the new data. Just 3.4% of sightings in the February 2017 data contains a specific notation indicating a near miss.”

Here are some of the AMA’s reports if you want to review:

(9/14/2015) – A Closer Look at the FAA’s Drone Data
(6/1/2016) – An Analysis of the FAA’s March 2016 UAS Sightings
(5/10/2017) – As drone sales soar, vast majority of reports remain simple sightings

The FAA’s Response to the Data Criticism

The FAA even commented on the data in the small unmanned aircraft registration rule, “Some commenters specifically found fault with FAA’s reliance on increased number of UAS ‘incidents’ reported to the FAA by manned aircraft pilots. Several commenters noted that the AMA analyzed those reported ‘incidents’ and found that out of the 764 reported records, only 27 (or 3.5%) were identified as a near mid-air collision, with nearly all of those involving government-authorized military drones. The commenters noted that most of the ‘incidents’ have merely been sightings of UAS. One individual pointed out that the FAA has published no analysis of its own ‘sightings’ data; nor has it disputed the AMA’s analysis of that data. This individual also asserted that a doubling in the rate of UAS ‘sightings’ in 2015 is consistent with the rate of growth of consumer small UAS, and is not cause for overreaction.”

Unfortunately, the FAA continued to press forward with just stating the raw sightings numbers.

Michael Huerta, the head of the FAA, said at Interdrone in September 2017, “We’re receiving an average of about 200 drone-sighting reports from pilots each month this year.”

The FAA’s Inaccurate Reporting of the Drone Sighting Numbers

Look at the data below for the total monthly drone sightings and compare that to the 200 number.

The average for January through August of 2017 was 190 sightings, not 200. If we say “about 200” and 190 are fine, which some may say is acceptable since the word “about” was used, you’ll notice that the average number of reported sightings drops if you shift the period forwards or backwards.

Notice that Administrator Huerta mentioned the number in September 2017 at Interdrone which is at the tail end of the active period. If he would have made the statement the next month in October, the average would have been 188.6. The average would have continued to decrease for the next several months. The 6 month average from October to March was 133.5.

On 10/12/2017, the FAA requested a review by OIRA to collect information for their LAANC system. The FAA said, “Today there are an average of 250 safety reports a month, or approximately 1,500 over a six-month period, associated with a potential risk of an incident between manned aircraft and a UAS.”

The big problem with this statement was the average was NOT 250! This is misleading. Just look at the graph and it is obvious that it cannot be 250.

In the preceding 6 months, the total count (1,295) was LESS than 1,500 and would be LESS than 1,500 in the next 6 months based on previous data trends.

In other words, simple math shows the real average and real total were both lower than what the FAA reported.

We could allow some flexibility to former Administrator Huerta since he used “about” and was reporting at the end of the growth period but the FAA’s statement to OIRA about the average and the 6 month total were both completely wrong.

To give you context, during this period of time the FAA was creating regulations illegally and large numbers of sightings benefited the FAA’s argument that they needed to by-pass the laws governing the FAA in how they create regulations and the FAA is allowed to create emergency regulations due to the rapidly growing sighting numbers.

And please keep in mind these are sightings…..not mid-air collisions, near mid-air collisions, near misses, etc. They are sightings which DO include some near mid-air collisions. There have been a few mid-air collisions but they are not counted in this data.

So with all of that in mind, let’s see how the news media followed the FAA’s lead on the drone sightings data.

News Media Inaccurately Reporting the Data

Fueled by statements from the FAA , you can easily predict the media would follow step.

I’m going to embolden some key words. Study the date of publishing and the title of the article. Everything in parentheticals is my commentary.

November 26, 2014 – Wall Street Journal article originally titled “FAA Reports More Aircraft-Drone Near Misses” (Which is not true. Sightings were reported, not near misses.)
November 26, 2014 – Washington Post article titled “Near-collisions between drones, airliners surge, new FAA reports show” (Wrong. Sightings, not near collisions.)
August 13, 2015 – USA Today article titled “Drone sightings on pace to quadruple this year”
August 12, 2015 – FAA news article titled “Pilot Reports of Close Calls With Drones Soar in 2015″ (Wrong. Sightings, not close calls.)

It’s interesting to notice how the media, and FAA, even hyped the month of over month percentage increase during the very active periods (April – September). The USA Today article said “quadrupled” but in reality the data was showing an increase from 238 total for 2014 to 650 for January through August 2015. Percentages lead to shocking titles while mere numbers tend to be rather boring. Would anyone really care if 500+ more drones were seen in the entire United States where millions and millions of people live?

No.

Interestingly, the data showing a decrease in drone sightings has not been reported by the FAA or the news media. The data shows that 2019 had a 6.76% drop in sightings from 2018.

The drone industry forecasts and projections all seemed to be doing the same playbook as real estate prices during 2004-2007 with endless growth. At a certain point, the market will be saturated.

I don’t know why the drop. This is before the whole COVID-19 situation so we can rule that out. Perhaps it could be because everyone purchased a drone in 2015-2018. Why would they want to upgrade to another drone since the one in their closet still works? I suspect it is like iPhones, you really don’t upgrade unless you really need the newer one or you broke your current one.

Graphs Created from the Drone Sightings Report:

These totals and graphs are based on data from 11/13/14 to 3/30/2020.

drone-sightings-versus-drone-registrations

How You Can Use the Data:

Everyone

Consider the graph above showing medium to large animal impacts on manned aircraft versus drone sightings. It seems that we need a more balanced approach to mitigating all risks, including large and medium animals which happen at higher numbers than sightings. Please don’t miss this point. Animal impacts are higher which means animal sightings are way way higher than drone sightings. Those animal impacts resulted in 24 people being injured but no fatalities. I’m not discounting the bad drones, just bringing them into proportion with the rest of the risks manned aircraft face.

Drone Operators

If you are operating near large populations, during the day, or during warmer months, you have a greater chance of having problems with local law enforcement misidentifying you as the other reckless crazy guy flying. Additionally, if you live near a large population that has a warm climate, you have a greater chance of having more state drone laws and/or local laws being created to counter the more numerous drone sightings.

If you are flying near cities, I HIGHLY HIGHLY suggest you Plastidip your drone some weird color so you don’t look like all the other bad drones. Think about it. You could easily get blamed for the other person’s illegal flight. Just spray your drone some weird unique color. You know those chocolate mint candy canes that are all green and brown. Do that to your drone. Ain’t nobody gonna be doing that. Well….unless they read this article. Nevermind.

Manned Pilots

Drones typically fly around populated areas during warm weather. Take this into account when doing safety risk management. Maybe try to fly higher if you are flying over heavily populated areas during warmer periods of the year.

Cities & States

If you don’t have that much of a population, you should think twice about creating a drone law as it will impact businesses greater than it will benefit the public which is small. Conversely, if you are a large city or heavily populated state, you should seriously consider figuring out a legal game plan for dealing with the drones.

Important Thoughts on the Drone Sightings Data:

The FAA has not “cleaned” the data from all lawful flights.

What you end with is this large number that is thrown out all over the place on the news. Many think the total number is the actual number of drone near misses because they don’t really bother to look further into the data. Many of the drone sightings are simply harmless sightings that could be drones being lawfully flown.

This is an extremely important point. The sightings will have two groups in it: the lawful and unlawful.

The FAA gave us some big numbers without indicating how many of these “sightings” were lawful or not. They didn’t “clean” the data for lawful flights. The May 2018 U.S. Government Accountability Office report said, “FAA also told us that some of the reports, despite the reporting pilots’ concerns, may have involved UAS operating in a safe and authorized manner.”

So have there been many lawful flights near airports?

Yes, many.

The FAA’s Federal Register indicated that the FAA “has processed 14,334 and continues to have over 6,000 authorizations in the processing queue.” Those are COAs prior to LAANC.

On 12/2/2019, the FAA reported,”To date, there have been more than 170,000 approved authorizations through LAANC.”

At the 2019 FAA Symposium, a powerpoint slide said that there have been over 37,000 manual authorizations granted and 87,000 automated authorizations granted.

This gives us at least 207,000 lawful drone flights near airports. There could have been multiple flights under many of those COAs during their duration.

How many of those do you think made it into the drone sightings data?

The FAA has not made the drone sightings data easy to work with.

Before you read my points below, compare it to the FAA’s Aviation Safety Information Analysis and Sharing portal. Just search around a little bit and you’ll see how just in-depth and detailed the data is. You’ll also notice that it is very searchable. Just compare these two screenshots to see what I mean.

FAA’s Excel Spreadsheet of Drone Sightings

FAA’s Aviation Safety Information Analysis and Sharing

I’ll give you some examples of why I say the data is not easy to work with.

The quarters do not have consistent formats which means you have to rearrange things. You’re going to have to do copy and pasting and use the left, right, or mid extraction function a bunch. Even that doesn’t work so well because of the terms and format change.
A whole quarter has an apostrophe in front of EVERY date which prevents you from sorting and totaling by date unless you remove it.
The data in the preliminary reports are all in one chunk for each line so you have to use more advanced Excel extraction formulas to extract the text which prevents lay people from analyzing the data.
The dates have times included in the dates which create more problems for sorting. The time isn’t extracted to its own column. You have to use the text to column function to fix this.

Frequently Asked Questions

What does this drone sighting data tell us?

It tells us that there are less reports of drones being seen.

How many of these drone sightings are of bad illegal drones?

We cannot tell if the drones were bad, but we can tell if they were a safety risk. Academy of Model Aeronautics reported that somewhere between 3.3 – 3.5% of the sightings included the term “near mid-air collision” and the Unmanned Aircraft Safety Team reported that from November 2015 to March 2017 “only a small percentage of drone reports” posed a safety risk. 96% of the sightings are just mere sightings.

Where do the drone sightings typically happen?

The sightings happen near airports as well as away from airports. The sightings also happen more frequently during warmer periods, in warmer environments, and over dense populations.

Are drone sightings going up?

No, they are decreasing overall.

How many of these drone sightings of drones appearing to pose safety risks actually turn out to be mid-air collisions?

Very few. Of the small percentage of drones posing safety risks, only a small portion of them actually end up hitting an aircraft which results in damage to the manned aircraft ranging from minor scratches and dents to requiring the replacement or repair of an expensive portion of the aircraft costing $100,000+.