My (Self Propelled) Movements Around Seattle
On February 16, 2015, I decided to start tracking all of my self-propelled movements (ie. no car, no bus). This is the resulting map of my traversed routes, overlaid on a map of Seattle. I separated the routes based on what I walked (blue), jogged/ran (red), biked (green), kayaked (purple), or played soccer (orange). I’ve blocked off the area I live in for privacy, though I’ve annotated both my and Anna’s places of work (UW and downtown, respectively) with a black dot.
Here’s the same map, but zoomed in to see more detail for the most traversed region:
It’s been great for a number of things, such as inspiring me to explore new areas of the city (ranging from exploring far away neighborhoods, to keeping track of new routes I could take for my commute to/from work). It’s also helped me be more active in general (I’d oftentimes choose to walk/run and add to the map, rather than commute by bus and lose those potential data points).
Furthermore, as you have likely noticed, I used my gps-tracking hobby as part of my three-part engagement proposal to my now fiancée. A zoomed in map, as well as a little bit of additional information, can be found here.
Details on how I made the map: I tracked my gps movements using my iPhone (initially using the Runkeeper app, though I later switched to Strava) or a smaller gps tracker I purchased from Amazon for ~$45 (used for tracking kayaking and soccer). I put together a script in R that imported the resulting gpx files and parsed them for latitude, longitude, and time information (Using the R package, “xml”). I added a section of the script that projected intermediate data points anytime there was a 2-5 second break between logged points, for consistency between individual tracked routes on the iPhone (which varied a bit in logging frequency, while the smaller gps logging device always collected data every second). The “ggmap” package in R was used to retrieve google maps at the various zoom levels, and I overlaid the gps data points using the “ggplot2” package.