Elevation Map of Tweets

Caitlin Dempsey


Twitter has release new maps to highlight just how prolific tweets are in London, San Francisco, and Istanbul.  Building off of the Geography of Tweets project that mapped out billions of geotagged tweets since 2009 (when Twitter started to geolocate them), this new mapping project takes a topographic look at the density of all geolocated tweets ever recorded in those three cities.    The maps were created by Nicolas Belmonte, a data visualization scientist with Twitter.  The project is named “Andes” and asks, “What would happen if all geolocated tweets in San Francisco were converted to an elevation map?”

To view the 3D map of geolocated tweets, you will need a browser such as Chrome or FireFox that supports WebGL (much like Google MapsGL).  If you access the project from other browsers such as Safari, you will be presented with a video demonstration of the maps.  The map tiles can be symbolized with a variety of choices from topo lines to Stamen’s watercolor maps (my favorite). The 3D maps is meant to be explored without roads or place names but if you absolutely want to see those, choose the “dark” view which presents a black and white map overlay with major street names.  The intensity of the elevation spikes can be changes from low to medium to high.   Use the trackpad  or your mouse to pan and tilt the map.  Select one of the three cities available from the drop down to toggle between them.


Photo of author
About the author
Caitlin Dempsey
Caitlin Dempsey is the editor of Geography Realm and holds a master's degree in Geography from UCLA as well as a Master of Library and Information Science (MLIS) from SJSU.

Free weekly newsletter

Fill out your e-mail address to receive our newsletter!

1 thought on “Elevation Map of Tweets”

  1. It is very interesting visualization.
    I want to use the original data for analyzing the relation between number of tweets and the economic activity.
    Could you tell me how I can access to the original data. Even aggregated data is enough
    for my purpose as long as it is georeferenced?

Comments are closed.