Georeferencing Century-Old Fire Insurance Maps

So recently I discovered these old fire insurance maps from 1914 of my hometown of Paris, Texas.

I was pretty impressed with the level of detail they conveyed. By referencing the following Sanborn key/legend it was fairly straightforward to determine specific traits of these older buildings / houses such as the materials used and the number of stories.

I've always been a bit curious about what the town of Paris looked like over a century ago and these insurance maps were particularly fascinating to me since they were published just two years before an actual devastating fire had destroyed much of its central business district. So in effect those documents now serve as important historic records.

Note: Ironically on that same day in which this blaze started, many of the city's leaders had met earlier to discuss a city plan proposal to renovate and extend the city. Ultimately those plans were dismissed though I do wonder if they would have been accepted if the fire's damage and subsequent rebuild costs had not refocused priorities.

Gazing at each of these insurance map sections individually was fine for casual curiosity but I figured it would be more convenient and practical to stitch them together into one larger map. So I did just that and used a current city map image as a guide layer to help me accurately align the pieces. Afterwards I used Georeferencer to convert the resulting image into an official map overlay so that I could then easily compare it against a modern city map from today ( using a slider to adjust the transparency. )

You can click here to see the final result. Many of the old street names have changed since long ago and I found it much easier to identify them this way. Plus it was a pretty neat way to see the overall road changes and city expansion across 108 years.

Today there are over 35,000 Sanborn insurance maps hosted online at the Library of Congress website. If you're inspired to create your own digital time capsule then you may find your own hometown there as well. This link covers more detail on the Sanborn map collection including additional details on how to read them.

If this interests you I recommend checking out Justin B. Sorensen's Utah project as well. He took the idea even further by generating 3D models from the insurance data.