OKC Memorial in 2016
I run…well, I attempt to run, but it’s fun for me! There are plenty of posts that walk through how to map out runs. Catalina marathon, Runkeeper routes, and the one that got me hooked are some of the ones that come to mind.
I played around with the entire dataset of my runs a while back:
However, the Catalina marathon blog post inspired me to start visualizing my individual runs.
I started with a fundamental thought of “automated” as much as possible. Starting with exporting a run from Strava, Runkeeper, or your tracker of choice as a GPX file all the way to viewing the dataset on a map.
Tulsa Route-66 in 2014
Using togeojson, I convert the GPX file a geojson format of the run. When uploading the tileset to Mapbox, I did not have much control over the metadata (elevation, heart rate, segment time, etc.). The output of togeojson are parallel arrays: one of the arrays is the linestring of the route and the other arrays are the metadata sitting in the properties object of the feature. In comes Turf to help out.
Using Turf meta, I merged the parallel arrays to an array of objects (datapoints).
I create segments of the run by grabbing the n and n+1 datapoints. I wanted to minimize overlapping segments, so I used turf.distance and turf.lineSliceAlong to crop out 2 feet of each segment. After that, all the segments are stuffed into a feature collection.
Walt Disney World in 2016
With the metadata of the runs attached to each segment, I was able to take advantage of some great gl styling features. I used data driven styling to visualize my heart rate using the same heartrate colorization of Strava. I also used 3D extrusions to visualize the elevation of the run, but mapping my Oklahoma runs will not show much.