After having seen the many iterations of the global connectivity map, I have decided to give it a try myself. The maps are being produced using this data, which offers public air-traffic data in tabular format. The data is available in three categories: airports, airlines and routes. The airports and routes data can be manipulated, combined and cleaned-up to create a complimentary table that lists the longitude and latitude of each flights origin and destination. This allows us to connect each flight origin and destination. However, straight lines of connectivity would create a messy and aesthetically appalling map. By using great arc circles as our method for creating the aforementioned lines of connectivity, we can create an atypical flight route map.
The map has been created in ArcMap using the symbol levels technique. This allows us to create a hierarchy of symbol levels such that the longer more obtrusive flight paths are drawn below the shorter flights paths. I’ve also taken the liberty to customize the colouring scheme such that the longer flights are drawn in a less prominent (dark)red. As the flights become shorter in length you will notice that they also become lighter in hue. This helps separate the 14,000+ paths of flights from
January of 2008 the database that was created many years ago (note: the map title and the original posting was stated incorrectly).
Full size available here: click me!
The openflights data package contains more than 50,000 flight paths
for January 2008 since the database was created, however I had trouble manipulating the data to produce lat/long for each origin and destination. Therefore, I am left with a sample of the total routes to display on the map. Notwithstanding the setback, I believe that the map has produced an effective result.
I should also note that even though this map uses a sample subset of the total number of flights, we can still clearly make out many of the continents and major regions around the world! I find it very neat that its possible to make out the general shapes without any underlying base map of the world with just a sample of the flights.
As mentioned early, there are many iterations of the global connectivity map. Some use the openflights data and some use various other sources. An example of other sources of global connectivity is shown by James Cheshire, who posted a vary nifty map, which display a sample of the 2007 global migration patterns.
I haven’t decided if I want to write a tutorial that would reproduce this map because there are plenty of sources out there at the moment which discuss the technique. Please feel free to visit Paul Butler’s page which shows the ever-so-famous Facebook Friends connectivity map. Also there is great documentation and discussion on how to use R Project for Statistical Computing to create a global connectivity map using great arc circles here.