Old maps are beautiful and useful, but they are just pictures until they are georeferenced, which means assigning them real-world coordinates so they line up with modern data. Georeferencing an old map in QGIS takes an hour, and the result is a historical layer you can compare with the present day. This guide walks through the whole process, step by step.

What Georeferencing Does

Georeferencing takes a scanned image of a map and warps it so that points on the image match known coordinates on the Earth. You pick control points, features visible on both the old map and a modern reference, such as road junctions, river bends, or church locations, and the software calculates a transformation that aligns the image. The result is a raster layer that sits correctly on top of modern data, so you can compare past and present directly.

Preparing the Scan

Start with the best scan you can get: at least 300 dpi, saved as a TIFF or a high-quality JPEG, with the whole map sheet visible and no skew. If the map is folded or torn, flatten it before scanning, because creases distort the geometry. Note the map's projection and datum if they are printed on it, because that information helps you choose the right transformation. A good scan saves an hour of frustration later.

Picking Control Points

Open the Georeferencer in QGIS, load the scanned map, and add a modern reference layer underneath, such as OpenStreetMap. Click a feature on the old map, then click the same feature on the reference, and repeat for at least six points spread across the whole map. Spread matters: points clustered in one corner produce a poor transformation everywhere else. Use features that have not moved, such as churches, bridges and river confluences, and avoid features that have changed, such as roads that were rerouted.

Choosing the Transformation

For a map that is already roughly accurate, a Helmert or polynomial transformation of first order is enough. For a map with local distortions, such as a hand-drawn survey, a thin plate spline transformation fits better but can distort areas far from control points. Start with a first-order polynomial, check the residual errors, and only move to a higher order if the errors are large. Residuals under a few metres are good; residuals over ten metres mean your control points need work.

Checking and Using the Result

After georeferencing, load the result alongside modern data and check that features line up: roads should follow roads, rivers should follow rivers. If a region is off, add more control points there and re-run the transformation. Save the result as a GeoTIFF, which stores the coordinates inside the file, and share it with the project that needs it. A georeferenced old map is a time machine: it shows what a place was, overlaid on what it is.

Old Maps and Historical Research

A georeferenced old map is a research tool: it lets you compare a landscape today with the same landscape a century ago, revealing changes in land use, settlement, and infrastructure. Historians use georeferenced maps to study urban growth, environmental change, and the impact of wars. The process is the same as any georeferencing project, but the payoff is historical insight, not just a map. The old map is a window into the past, and the georeferencing is what opens it.

Key Takeaways

- Georeferencing assigns real-world coordinates to a scanned map image - Use at least six control points spread across the whole map - Churches, bridges and river bends make the best control points - Start with a first-order polynomial and check the residual errors

Frequently Asked Questions

How do I georeference a map in QGIS? Open the Georeferencer, load the scan, and click matching features on the old map and a modern reference layer. Six or more spread-out control points produce a good transformation.

An old map becomes a dataset when it is georeferenced. Scan it well, pick good control points, and compare the past with the present.