Analysis of Dulcert 1339 Portolan Map

The whole map

With coordinates registered with Google Earth


See also previous blog Portolaan

Tyrrhenian Sea

Fit with Compass Charting:

In the graph of the local metric (i.e. how many kilometres there are in one pixel size) we see a large peak in the Eastern Algerian coast. That means that that the map is compressed there relative to the average.

So when we go around the Tyrrhenian Sea, mapping with a Compass Charting method, taking this place as start and endpoint would offer a good chance of giving a good fit.

Compass Charting is moving from place to place, and mapping each next place relative to the previous place according to the direction and distance of a constant compass course, without taking into account the curvature of the Earth. Note that following a constant compass course is not a straight line (or more precisely, a great circle). And when charting a large route with places around a central point, your map will not return to the same place where it started, because the circumference of a circle on a globe is less than 2π times the radius, while the length of your mapped course will be 2π times the radius, because you use the same scale everywhere.

And indeed, the Compass Charting method results in a better fit than the Mercator Projection.

With an average error of 22.8 km the quality of the mapping is very good. The Mercator projection has an average error of 26.0 km.

Going around with the cut in Algeria gives a better result than going around with the cut between Sicily and Algeria. That suggests that Tunisia was charted from Sicily and Morocco from Tarifa, with a joint in Algeria.

Atlantic Coast

Fit with Compass Charting, the average error is 25.6 km:

Fit with Mercator Projection below. Evidently the enlargement towards the North, that is typical for a Mercator Projection, doesn’t fit this map. The average error is 36.6 km.

This is the local metric of the fitted Mercator Projection:

And this is the actual local metric, calculated by fitting the real positions of a small neighbourhood of each place (note: the scale differs from the previous graph).

There is no systematic trend from North to South, just peaks caused by local distortions in the map.

3 thoughts on “Analysis of Dulcert 1339 Portolan Map

  1. I came across your blog because I found something elsewhere, that piqued my interest.
    In the Hapgood book, there is an illustration of the outline of the Dulcert Portolan that reads starting point at intersection of 30 degrees North and 5 degrees West; I could not determine how that was established from the book’s contents, but I checked Google Earth and found ruins at 29 Deg 32.50 N and 4 Deg 59.54 W, which is close to the given coordinates. Would this imply that this is the place where the map was copied from an earlier version?

  2. After analysing the map and trying out numerous projections, I came to the conclusion that the map is not based on geographical coördinates, but on compass directions. The greatest accuracy is in the area of Genoa, and generally along their trade routes.

    From the little that I read about Hapgood, I regard him as a crackpot, and I am not going to waste time reading his work.

    The best simulations today about the changes in the earth magnetic field centuries ago are not accurate enough to attribute the changes in the magnetic North on the Dulcert map from East to West directly to the real physical situation, but they don’t contradict them either.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s