This is the first in a series of blog posts reporting on techniques I worked on over the summer, as part of some work I did on the GEOIDE project, Visualizing Urban Futures. The project specialized in the visualization of transportation data, specifically origin and destination data (OD). This first post will cover choropleth mapping. Choropleth maps are used universally for zone-based data and can be used to visualize movements to and from areas.
For many jurisdictions, trips are coded and aggregated into zone systems. A jurisdiction may have access to a table that list the number of trips that are observed between each origin to every possible destination. Having a zone system (ex. as an Esri shapefile), the origin or destination can act as the basis for geocoding trips.
Since Choropleth maps can only show trips related to one region, it is essential to isolate trips that all originate from or are destined to a single area. The area can be a single area or multiple areas aggregated. The picture below shows some combinations of possibilities for choropleths. here, the blue zones are the origins of interest, while the orange zones are all the destination zones related to the single origin.
These sets of maps can be used for visualizing the major origins or destinations of groups of people. When combined with street networks, transit networks and major points of interest, viewers can begin to reason why people move to certain areas as opposed to others.
Case Study: Trips to Toronto’s Financial District
The maps below show a choropleth of trips from the area around Toronto to a single zone where much of the city’s financial offices are concentrated. Since the destination is the financial district, it can be assumed that trips destined for this zone will be for work, and that the workers will have relatively high incomes.
This first map shows a regional view of the Greater Toronto Area, with the financial district zone as a white dot. Here the direction of movement is indicated by the dashed arrow. it can be seen at this level that trips originate mostly in the area surrounding the financial district:
Upon zooming in, it can be seen that certain zones produce more trips to the financial district. For example, the first labeled zone (1) is the Pearson International Airport. It can be envisioned that people going to the financial district are travelling from international locations. The second zone (2) is an area of Toronto with a large number of households with a high median income.
Obvious problems with choropleth maps are the reliance on borders and the visually complex images they produce. With smaller zones, and a greater number of colours and borders, there are issues with visual confusion. One way to get past this is to use the largest, most regular zone systems that are feasible, and to minimize the number colours to less than 5, or greater than 20. in the first case, differentiation between zones is easier, while in the second, general trends on a larger scale can be seen. For a more nuanced understanding of issues with choropleths, check out this discussion at vis4 (Link)
The Moral of the Story:
Choropleth maps are an invaluable tool for quickly visualizing trip data, showing trends and rationalizing flows of people between zones at regional and local scales.