Switzerland is not exactly a country of major earthquakes. Nevertheless, in recent history earthquakes have happened, which – if they happened today – would cause considerable damage. The Swiss Earthquake Catalogue contains information on past earthquakes, their occurrence and consequences, in various regions of Switzerland. All historically known earthquakes are included in this databank. The Swiss Seismological Service (SED) (1), at home at the ETH Hönggerberg in Zurich, recently published a new version of the catalogue (2).

Deficient Database

"The previous earthquake catalogue was published in the 1970s", says Donat Fäh of SED, "and at the time the work was based on previous compilations, therefore on reports from people who had already set up similar lists at some time or other." The defect in this method is clear: "Any mistake in the compilation is perpetuated."

After the decision had been taken in the 1990s to revise the entire catalogue, the editors took the opportunity of critically assessing the existing catalogue in order to weed out mistakes. This could only be achieved within a sizeable project. A lucky coincidence helped to provide some of the necessary funds: Switzerland's nuclear energy plants were called upon to carry out a new assessment of the danger of earthquakes and they agreed to help finance the project. In all, around 20 people, mostly seismologists and historians, worked on the new catalogue.

Distribution of historical earthquakes, which have caused damage. large

Only easily accessible archives evaluated

In one part of the project, handed-down reports on past earthquakes were newly assessed. "For reasons of time and money we were only able to carry out thorough research in easily accessible, large archives", explains Fäh. "Research into local archives or church records was only possible in a very limited way. It is very likely that such sources could yield more information." Above all, there is relatively little information to be found on historical occurrences in the main archives in canton Wallis. Regrettably, as Fäh says, in view of the fact that this is one of the most important regions for earthquakes in Switzerland.

The catalogue has certainly profited from the work of historians, who were able to correct a number of errors. "We realised that many events had been duplicated in the old catalogue," says Fäh. Especially with regard to older earthquakes, which, owing to the change in the calendar in the 16th century led to a number of double entries. Moreover, mistakes happened with the association of locations or place names, and some are the result of transcription errors. In the old catalogue, for example, the earthquake in Lindau in 1720 was listed as serious because the compilations had described "destroyed houses". Historians found that this event was said to have produced "shaken houses" in the original reports.

Church Service in a provisional tent camp in Visp after the earthquake of 1855 (contemporary illustration). large

School children admiring a crack in the wall after the earthquake in Villeneuve (1946). large

Similarity of eyewitness testimonies

As important as original documents are, they too must be critically approached. Under the immediate impression of the event, eyewitness reports are often grossly exaggerated. Sometimes testimonies are clearly copied from other sources. Fäh tells of an occurrence that appears in many reports of a certain period: "Chroniclers report that during a certain earthquake a women threw herself out of a window with her newborn baby in her arms. Apparently this was one of the most terrible tragedies that one could imagine."

In all, more than 600 events were historically re-evaluated. 177 of them reached an intensity of VI or higher (3), which means they damaged buildings, sometimes massively. 260 events listed in the old catalogue were shown to be incorrect, at least partially, or were duplicates. Historians even discovered a few earthquakes that had not been known about until now.

Conversion to magnitude

In a second step the historically passed on data were seismologically interpreted. All earthquakes that had produced any effect (i.e. with an intensity of V or VI) were assigned to a so-called intensity field. These cards show what damage was ascertained and how strongly the earthquakes were judged to be by the people who experienced them. Finally, in a third step, seismologists calculated physical parameters for earthquakes for which they had enough data to determine the intensity field, such as the location of the epicentre, the depth class and the magnitude of the earthquake.

As no instrumental measurements exist for historical earthquakes, a calibration had to be developed to determine their intensity. "We evaluated data on the bigger earthquakes of the 20th century and compared them with the observed intensities", Fäh explains. According to this calibration it was possible to uniformly calculate the magnitude of historical earthquakes. In the near future the SED now wants to complete a new probabilistic seismological danger map for Switzerland based on results of these calculated earthquake intensities. The Swiss National Science Foundation recently granted a project, which will enable the historical research to continue.