Where the aliens are flying their UFOs

I came across some UFO sightings data on Infochimps, from the National UFO Reporting Center, and it seemed like a good excuse to mess around in R. I was just playing around, but the globular result was kind of fun to look at, so here it is.

The dataset is 60,000 sightings, but the above shows about 45,000 locations that could be geocoded immediately. The whiter the region, the more sightings there were in the area from 1906 to 2007.

Is it just me, or does the map above match up with this other map of major US airports?

Could some of the sightings actually be… airplanes? Nah. The aliens are coming. Luckily I have a big bat, several glasses of water, and asthma to keep me safe.

35 Comments

  • Or just population densities?

  • Jamie Bull July 7, 2011 at 3:58 am

    It might also have something to do with population density. It does seem to be a very close match though!

    • Merely assuming a uniform probability of any given person reporting having ‘seen’ a UFO would produce the same results, no data needed ;).

  • Jamie Bull July 7, 2011 at 3:59 am

    Beat me to it…

  • Simon Jones July 7, 2011 at 4:10 am

    If I had do land my UFO on a strange planet, I’d try their airports first.

  • not sure how the data set looks but if you did some quick division by population… i wonder if it would look different in a sightings per resident view? Cool stuff.

  • Looks a bit like a honeycomb pattern, doesn’t it? What algorithm do you use to create the heat map? Also, could you please highlight Area 51 for me?

  • Agree w Pete C. Interesting premise but suspect airports act as proxy for population hotspots (the good old correlation vs. causation argument).

    Could pick almost any people-based metric and see a similar pattern… I’m thinking, universities? http://www.usjournal.com/en/students/help/map.php

  • Yepp, if you overlay a population related map the highlights match up pretty well. Would be interesting though, to see which points stand out if the ufo-map gets normalized by population =)

  • Hingehead July 7, 2011 at 5:20 am

    How come the number of UFO stories that make the news is inversely proportional the ubiquity of devices with video recording ability?

  • There you have it. Absolute proof.

  • Jean Wood July 7, 2011 at 9:25 am

    I found this veerrryyy interesting. I agree with the earlier posts about airports and population. It must have been fun for you to actually think about this and then do something with it. Please keep us all alerted with the latest information their coming. I’ve never met an alien I didn’t like…lol

  • Harvey Summers July 7, 2011 at 9:39 am

    The data is skewed since the cows quit reporting.

  • Christian July 7, 2011 at 9:41 am

    I would believe that the representation of “reported sun sightings” would be quite the same. Given the quota of ‘UFO believers’ is constant across the population.

  • It would also be interesting to see this with a 3rd variable of time. Is the distribution similar between historical activities:
    1900-1930 (before large scale commercial flight)
    1940-1945 (war)
    1955-1965 (Cold War)
    1990-today (when travel for the ‘ordinary’ business person became common)?
    2005-today (pretty much everyone in the country has access to the internet’s easy-to-create information)

    And comparing it to immigration (the terrestrial kind) might also be interesting. Do report frequencies increase when or where we get rapid increases of immigrants?

  • So we could potentially tease out the different effect of population centers from aircraft activity by looking to see if there are more reports in unpopulated areas in major air traffic routes than in unpopulated that are not in major air traffic routes. Looking at the map there appears to be some straight lines between major airports, which could be air traffic routes, but couple also be artifacts of interpolation. Which leads me to ask, what method of interpolation did you use?

  • luisferfranco July 7, 2011 at 12:12 pm

    Very interesnting.

    At first sight I tought it was a map representing democrat vote. So… if you map UFO sightinings to voters… what do you think you’ll get?

  • UFOs are also more common along LEY LINES (energy/dragon lines, where there is increased magnetism), and nuclear plants and radiation site !!!

  • I’m glad Canada is completely black…. I was worried for a second. :P

  • Way to ruin it, Nathan.

    I want to believe.

  • What if you made a map of UFO sightings by county divided by county population to see if certain locales are more UFO-dense?

    Another interesting analysis along the lines of your airport theory would be to take a dataset of known aircraft flight paths and create a distance (20 mile?) buffer around each of those lines, then sum the total number of buffers above a specific county/gridcell and then divide the total number of UFO sightings for that particular county/cell by the total number of likely airplane flightpaths visible from the ground.

    This would give you a by-density comparison of UFO sightings to likely airplane sightings and see if there is a correlation.

  • It’d be really cool if you could write some kind of tutorial on how to do this in R . Great work!!

    Matt

  • Is there any ohter country mapped?

  • Nathan, could you rerun the same analysis but discount for population? That way we’d be able to see places where there are exceptional numbers of sightings.

  • Let us also consider cheaper housing is closer to airports and those using cheaper housing are also more likely to engage in drug use, which then affects the probability of seeing “UFO’s.”