Crime Data Heatmaps

A heat map is a way of presenting data that makes it much easier to understand the density of a variable. Rather than just plotting each pont on a map, a point on a heatmap adds a bit of ‘temperature’ to a location, so a lot of points close together make an area “hotter”. There are several ways of constructing heatmaps, but they all generally represent the areal density of a variable, and when the viewer is most concerned with density, they are the best representation. For instance, here is a typical crime map, showing an icon for each crime incident:

Crime map from Crimedata.com
Crime map from Crimedata.com

While this map is useful for determining what crime incidents may have been recorded near a particular address, the size of the icons obscures the variation in incident density. Here is a map of the same area, from the Trulia Crime Maps website, presented as a heat map:

Heat map of Pacific Beach, from Trulia.com
Heat map of Pacific Beach, from Trulia.com

The heat map isn’t useful for getting information about a particular incident, but it is a much better representation of where crime is happening, in this case, clustered around bars and the beach, while few incident are occurring in the residential areas.

This crime heat map of London allows you to adjust the radius that each crime incident contributed to the map, as well of intensity of the event, so you can get a sense for how those parameters work.

If you’d like to generate your own heat maps, many GIS programs can create them, and heat maps are increasingly common in web-based mapping services. Here are a few online services you might try:

  • HeatCanvas, heatmap library for HTML5 canvas
  • gheat, an open source program that works with google maps.
  • PyHeat, a similar program for the Python language.
  • Open Heat Maps, a free web service.
  • Heat Map Tool, a commercial web service.
  • Heat Map Api, a commercial service, although it has a free tier.