Analyzing federal surveillance planes with CartoDB

Using CartoDB Deep Insights to analyze data from BuzzFeed News on federal surveillance planes

One week ago BuzzFeed News broke the story of U.S. government surveillance planes routinely circling over most major cities. It was while reading the Hacker News thread on the topic that I found the data, along with notes on the analysis, available on GitHub.

As I showed on my flights routes post we are working at CartoDB on a visualization solution to explore and analyze location data.

This time I didn't have to collect the data as BuzzFeed News had already done the job. I just recreated their analysis with a set of widgets for dynamic filtering.

What is @putohelicoptero looking for?

Checking the methodology used by BuzzFeed News I asked myself:

What if we could do the same analysis with Spanish cities?

Actually, this is something many Madrid citizens have already brought up before: If you live near Madrid city center, you'd have noticed the annoying noise of a helicopter flying over the city on some quiet nights. Just to give a rough idea, there's a Twitter account called @putohelicoptero featuring the helicopter.

Digging for information I found the model of the helicopter in a news article by El Mundo, reporting all the inconveniences that the now known EC-135 'Cóndor' caused.

As explained in the BuzzFeed News analysis, Flightradar24 receives data from a network of ground-based receivers, 1 you just need an identifier for each aircraft to access that data. If only there was an aircraft registry similar to the Federal Aviation Administration registration database I could identify the model the "Policia Nacional" flights.

And it exists.

The Registro de Matrícula de Aeronaves Civiles publishes a document with all the aircraft registered in Spain. 2 At first glance, at least 41 of those aircraft were a match with the model I was looking for, and then a colleague helped me introducing them in Flightradar24 for further filtering. 3

Once we found it, we just had to export the location data and map it in CartoDB. As an example, this is the flight of the helicopter for New Year's Eve, checking Santiago Bernabéu stadium (Start of San Silvestre Vallecana), Vallecas (End of San Silvestre Vallecana), and Plaza de Sol.

At a lower scale, "Policia Nacional" is also conducting surveillance flights that can be tracked, and later analyzed. At last we can already answer the question:

What is @putohelicoptero looking for?

We can see in which zones were conducted most of the surveillance flights, performing a spatial join to the boundaries of Madrid barrios (activate the layer in the map above). 4

Disclaimer: Flightradar24 has a pretty restrictive license, thus I haven't shared the data from @putohelicoptero. The data for federal surveillance flights is already available in BuzzFeed News GitHub account.

  1. You can build your own receiver with a Raspberry Pi and a low cost DVB-T USB stick, or apply for a receiver
  2. Last updated on 1st April 2016, as for the day of publication of this post
  3. You need a Premium account for more than 7 days historic data, but it's easy to obtain. Thanks to @javitonino who assisted me in this process.
  4. Data from November 2015 to March 2016.

Tweet this

Published in Carto. If you want to receive new articles when they come out, enter your email address here or subscribe to the RSS feed.