vis.gl
Published in

vis.gl

Using Kepler.gl and Movement to Visualize Traffic Effects of a Rainstorm

On January 12th, 2018, the Pittsburgh area experienced major flooding due to a severe rainstorm. Major roads and freeways were flooded as commuters sat through longer travel times than normal. By visualizing Uber Movement data in kepler.gl, Uber’s newly open source geospatial analysis tool, we can visualize the impact of the rainfall on by comparing travel time from downtown Pittsburgh to all zone destinations to the same day a week before.

Figure 1 below shows travel times from zones of 12 or more minutes from downtown. The left map shows data from January 12th, while the right map visualizes times from January 5th, 2018 as a baseline comparison. This post breaks down how this visualization was made using kepler.gl and Uber Movement.

Split screen visualization of Movement travel times data for Pittsburgh

Rainfall in Pittsburgh

Hover tooltip for travel times on January 12, the day of the rainstorm

On January 12th, commuters experienced a significant increase in travel time, as compared to the week before, especially on and around Interstate 376.

Hover tooltip for travel times for January 5th, the week preceding the rainstorm

kepler.gl’s side-by-side comparison feature allows us to see the two dates easily. Using the hover tooltip, we can see the difference in travel time to zone 549 is as much 97 minutes!

How Did We Visualize This?

Those who want to create their own comparative visualizations with kepler.gl and Uber Movement data need only follow a few simple steps:

  1. First, download the desired origin zone to all destination zone data from Movement (example here). Download this data set for all dates of interest.
  2. Second, download the “Geo Boundaries” GEOJson file for that city.
  3. Through your tool of choice, join these two Uber Movement datasets using the shared zone IDs included in both files, and append the data sets for each date of comparison. For example, in our final data set, we have one column for day1_geom, and another for day2_geom) and feed this dataset into kepler.gl.
  4. In kepler.gl, create two layers, one using day1_geom, the other using day2_geom. Make sure to show ‘quantize’ coloration. To better visualize the difference, you can use the filter feature to filter out zones that have similar travel times.
Filter pane in kepler.gl

5. Finally, to compare the two layers, you can use the dual map view on the top right of the screen, as shown below.

Dual map view selector in kepler.gl

Link to this kepler.gl configuration is available here for reference.

kepler.gl enables you to drive impact with location data. Hundreds of engineers, researchers, scientists, journalists, data scientists, GIS specialists, cartographers and visualization experts use it to understand, analyze and drive decisions about location intelligence. Today, Uber is opening this application to enable anyone to make a similar impact with their geospatial data.

--

--

--

Open-source, WebGL-powered visualization frameworks

Get the Medium app

A button that says 'Download on the App Store', and if clicked it will lead you to the iOS App store
A button that says 'Get it on, Google Play', and if clicked it will lead you to the Google Play store
Uber Movement Team

Uber Movement Team

Helping urban mobility through open data

More from Medium

Image Color Extraction using K-Means

How I built a wind map with MTS

Tutorial 8 — Querying Places Location API

Generate XYZ Tiles in QGIS and Serve on the Internet