The oldest portolan chart in the world seemed to appear out of nowhere at the end of the thirteenth century, and there are no signs of the existence of any simpler, earlier versions. The fact that these charts are so incredibly accurate despite the limited knowledge and measurement instruments has long been seen as a fortunate coincidence. In his research, Nicolai proves that even with the most optimistic assumptions this is, in fact, impossible. "The remarkable thing about his research is that he has studied portolan charts using insights and methods from modern-day geodesy. The results are utterly surprising", says his PhD supervisor, Professor Jan Hogendijk of Utrecht University.
So far, it has been assumed that sailors carefully recorded data about their courses and distances they travelled on busy shipping routes. It was believed that their measurements were compiled in so-called portolans, books with sailing instructions, and eventually processed by cartographers. Nicolai, however, has shown that it is unlikely that the nautical compass was available in time and that navigational methods used at that time were sophisticated enough to determine distances at such a degree of accuracy.
Furthermore, using averages in calculations to increase the accuracy of measurements did not become common practice until the eighteenth century. A mathematical analysis of the oldest surviving portolan has revealed that its source data must have been derived from a portolan chart - instead of the other way around.
Nicolai has also established that portolan charts were drawn on the Mercator projection, or a similar type of projection. A projection is a mathematical conversion of the geometry of the curvature of the Earth to a flat surface. There are no signs so far that indicate that the knowledge required for chart projections was already available in the Middle Ages. It is generally assumed that Mercator projection was not introduced until 1569 by its originator.