Historical Events and Observations

Historical geomagnetic observations from the Netherlands during the Carrington event (1859)

¹ Royal Netherlands Meteorological Institute (KNMI), Utrechtseweg 297, 3731 GA De Bilt, The Netherlands
² British Geological Survey, Research Ave South, Riccarton, EH14 4AP Edinburgh, United Kingdom

^* Corresponding author: kasper.van.dam@knmi.nl

Received: 6 August 2024
Accepted: 28 January 2025

Abstract

The Carrington event of September 1859 is the best known example of an extreme geomagnetic storm, often cited when discussing space weather risks for modern infrastructure. Historic observations including auroral sightings, magnetometer records and anecdotes of impacts on telegraph systems have been widely shared before, but none of these have included observations from the Netherlands. Geomagnetic observations taken in Utrecht and Den Helder during the Carrington event were digitised from the Royal Netherlands Meteorological Institute’s (KNMI) yearbook of 1859, and compared to much more detailed magnetograms from London. This combined analysis, beyond its application in communication with Dutch stakeholders, contributes to a better understanding of the interpretation, limitations, and uses of such archived measurements, of which more examples might be available in archives internationally. The observations consist of spot measurements taken three times per day. The Den Helder data only partially record the Carrington storm. Conversion factors from Den Helder have been used to estimate missing conversion factors of the Utrecht data. The correlation between the Dutch declination measurements and those made in London is strong with correlation coefficients larger than 0.7 for the Utrecht data and larger than 0.9 for the Den Helder data. However, there is very little correlation between the Dutch and British inclination measurements. The London horizontal intensity measurements compared to Den Helder data give correlation values larger than 0.8 but the observations from Utrecht match less well. There is a significant deviation between the British data and the Utrecht declination and horizontal intensity measurements during the quiet period between 30 August and 2 September. It is unclear what causes this deviation. Given the proximity of the locations and similarity in latitude, and based on the coherent registration of the measurements, it is reasonable to assume that the magnetic traces captured in London provide a good approximation of the magnetic field variations in the Netherlands during the storm, indicating that these may be used for impact assessment studies for Dutch vital infrastructure.