Have you ever wondered how historic maps can be used with today’s modern mapping technologies? One of the ways in which analog maps can be used with GIS (Geographic Information Systems) is through a process called georeferencing. Georeferencing is the process of adding digital spatial reference information to an otherwise non-spatial image. Adding spatial reference information to a scanned map image allows the map image to align correctly with the geographic features it was built to represent – which allows a user to layer any other spatial data file alongside (or on top of) their map image!
Today I’m going to give you a peek at how this process works, using the following 1967 map of the US Capitol grounds as an example.
This map shows properties under the jurisdiction of the Architect of the Capitol in 1967. Missing from this map is the Madison Building of the Library of Congress, home to the Geography & Map Division. It would be fun to compare this 1967 map to today’s Capitol Hill Complex to see how the area has changed over time. Let’s do it!
Maps that are scanned as image files meet the criteria for what is called raster data: data composed of a continuous grid of cells (or pixels). Because it is common for spatial data to be stored in a raster format, scanned map images can be loaded directly into GIS software without any file conversions needed. Geographic data layers are able to align correctly when viewed in GIS software because of what is termed “spatial reference information.” Without this spatial reference information, when you load a scanned map image into GIS software, the image will default to the coordinates 0°N 0°E (also known as Null Island). The process of georeferencing an image corrects this by allowing a user to manually add control points between the non-spatial scanned map image and a pre-existing GIS data layer. This GIS data layer can be anything that already has spatial reference information and displays correctly in GIS software.
To place a control point, the GIS user will take advantage of georeferencing tools provided by the software package they are using (georeferencing tools are available in all of the most widely used GIS software options out there). These tools should allow a user to place a control point by selecting a specific point on the scanned map image, and then selecting the exact same point on the GIS layer. Once the user adds a couple of control points, the scanned map image will begin to align with the existing data layer scale, but may not be in entirely the right geographic alignment yet. Here’s what my map looks like after placing two control points: one on the Capitol rotunda and the other on the West Front Fountain:
Though those two spots are aligned correctly, the rest of the map is not. A user must continue to add control points, making sure they are well-distributed across the map image, until it is determined that the two layers are aligned properly. Here’s what the map looks like after 21 control points have been placed:
It looks good! Once this is completed, the georeferencing can be saved, allowing the scanned map image to “go to the right place in the world” whenever it is loaded into GIS software. The map can now be displayed with other GIS layers to create new visualizations. Here is a look at the original map transposed against current aerial imagery:
In the lower right-hand corner of the map we can now see the current aerial footprint of the Madison Building where there was a vacant lot!
There are many reasons why a georeferencing a scanned image can be useful. The primary reason is that it allows a map user to view the map in geographic context with any number of other spatial data sources. This could allow a user to more directly compare maps created at different scales or in different times periods, or to see an older map juxtaposed against current aerial imagery or spatial data. It also allows a GIS user to use the scanned map image as a basis for the creation of new spatial datasets – more on that topic to come!