Mapping Data with QGIS

ride0185

We make a lot of maps at Fat Pencil Studio– or rather, we deal with a lot of geospatial information. When the area in question is relatively small (a city block or so) we usually turn to aerial imagery from NearMap or other providers, and add labels, arrows and other marks to show a sequence of events, the positions of people, vehicles, cameras or evidence. When a wider view is needed, we create base maps in MapBox studio or use topos from USGS. And once in a while, when we need more flexibility, we turn to Geographic Information System (GIS) software for a custom solution.

GIS software allows us to assemble many layers of geographic information, then combine the layers into a single map, not unlike Adobe Photoshop. Our mapping software of choice is QGIS – an open-source, free alternative to ArcGIS. Base maps can come from USGS, Mapbox, Open Street Maps or others.

Sometimes a basemap that highlights features of importance is all we need. In other cases, we might add additional data layers. In this case, we added dots to represent Oregon tobacco inspections. Here, we used a dataset from the city of Portland for a study of building age and height. We've also used city data to compare land value to parking rates. Below are a few other example layers, using data from Portland Maps, DOGAMI and others.

We can also plot time-based data: the location and speed of a police car from its onboard data recorder, cell tower pings, and more. Data with a time component can be exported as an animation. The example below uses a GPS track from a bike ride I took last summer on Forest Park's Leif Erickson Drive, recorded with the Ride With GPS app on my phone. (It's been sped up somewhat here!)

We can also use QGIS to work with historical photos or detailed lidar scan data:

map-photo-1939

A set of aerial photos from 1936, orthorectified to accurately match Portland's geography. Note the logging in Forest Park and airport still on Swan Island.

detail-highest-hit

A lidar scan captures the shape of buildings, cars and the tree canopy in Forest Park. Data like this can be processed in QGIS and converted into a 3d model of terrain– a process we use occasionally for projects where accurate terrain is necessary.

Not every case we work on requires a set of custom maps or a detailed terrain model, and we don't always have relevant location data for people or vehicles involved in a case. But when we do, QGIS gives us a powerful and flexible platform for organizing, analyzing and displaying geospatial data.

joel.jpg

Joel Newman is a Senior Designer at Fat Pencil Studio