Visualizing telemetry data for the Storm Area 51 event
By: Sofia Heisler
Of all the weird and wonderful news events of 2019, one of my favorites was when the internet banded together and decided it was finally time to find out the truth — about Area 51.
In early July an event appeared on Facebook calling on netizens to storm Area 51, a secretive U.S. government research facility and a long-time favorite of conspiracy theorists intent on proving that the government is hiding extraterrestrials from the rest of us. The goal was to break into the compound at 3 am on Friday, September 20th, to “see them aliens” by swarming the facilities via high-speed “Naruto” running. The thinking was that if a whole lot of people headed towards the high-security base at once, the military couldn’t stop all of them.
By the sheer power of its absurdity, the event became a viral sensation, creating a flurry of Internet memes, attracting over three million RSVPs, and, apparently, earning the organizer a personal visit from the FBI. But as most of us know from experience, an RSVP on Facebook is hardly a guarantee of attendance, so I started wondering — did anyone actually make it out there?
At Mapbox, we have a unique level of visibility into this question through telemetry data. We collect highly anonymized location information from mobile phones running applications that use Mapbox maps. We can use this information to understand traffic patterns, trends, and in this case, alien investigations. Given that all of our telemetry information is anonymized, I can’t tell you if any of the location events came from actual extraterrestrials; but I thought it would be fun to see if I could visualize the historic storming of Area 51 with our data.
Visualizing the event
To start, I had to actually find Area 51 on a map and see what roads were nearby. Interestingly, while roads leading up to the compound are clearly apparent on satellite imagery, on some maps the area shows up as blank wilderness. Fortunately, this wasn’t an issue with the Mapbox resources at my disposal, since Mapbox maps use data from over 130+ sources, including OpenStreetMap.
Once I had our map, I loaded up some information about traffic in the vicinity of Area 51, pictured below. The areas colored in red are closed to the public.
The video below shows the volume of traffic on the roads near Area 51 starting at midnight on Friday, September 20th. The traffic is colored by a number of observed cars (green being the lowest, red being the highest). Area 51 is in the lower left-hand corner; the closest place to get to it via a publicly accessible road appears to be near the town of Rachel (right near the “Red Zone” tag).
It looks like showing up for the attack at 3 am was a bit of a stretch for all but the most dedicated. In the early hours of Friday, September 20th, only a few cars can be seen on the roads near the compound; the bulk of the car traffic arrived on Saturday and Sunday afternoons. Of course, we’re only tracking moving car traffic; I’d like to believe that the mass Naruto run on the compound did happen, it just wasn’t captured in this particular dataset.
But how did that weekend’s traffic compare to the normal traffic in the area? To answer that, I created a time-series chart of daily car traffic in this area, based on our telemetry data (below). On the weekend of September 20th, the traffic spiked to more than double the average of the rest of the summer. Someone did, in fact, make it out just to see them aliens!
The Storm Area 51 event seems to have boosted the facility’s popularity as a tourist destination in the months that followed. But interestingly, the traffic chart also shows that there were two other times in 2019 when Area 51 reached a similar level of popularity: one was on Halloween, and the other on May 4th, also known as Star Wars day.
Even if the storming was only hype, I’m still holding out hope that by the powers of the internet combined, we will one day uncover the secrets of Area 51. Want to be around to track the traffic and the aliens* when it happens? Or just interested in working on a platform that powers over 45,000 apps, touched by over 600 million people each month, and processes data at a petabyte-scale? Apply to work with my team at Mapbox!
* No guarantees about the aliens. Sorry.
Sofia Heisler - Data Scientist - Mapbox | LinkedIn
Did we find aliens in 2019? was originally published in Points of interest on Medium, where people are continuing the conversation by highlighting and responding to this story.