October 8, 1909

1909 Kupa Valley earthquake

Kupa Valley earthquake produces the records behind Mohorovičić’s later discovery

Mohorovičić did not announce a boundary inside the Earth on the day the ground shook. He posted 430 questionnaires and published a year later.

Thirty kilometres south of Zagreb

On 8 October 1909 a damaging earthquake struck the Kupa Valley in the Pokupsko region, roughly 30 kilometres south of Zagreb. It was felt over a radius greater than 200 kilometres.

Two people were killed and many were injured. A mason died under a collapsing church roof at Lasinja and a roofer fell from a bell tower at Šišinac; the only damage photograph centennial investigators found shows the collapsed front wall of a church in Sisak.

Mechanical seismographs in Zagreb and across Europe recorded the earthquake's P- and S-wave arrivals, creating the observational record Andrija Mohorovičić went on to analyse.

430 questionnaires and 41 stations

Mohorovičić, director of the Zagreb Meteorological Observatory, immediately circulated macroseismic questionnaires, joined a government investigation and visited the epicentral area on 12–13 October.

He gathered 430 questionnaires and arrival-time data from 41 seismic stations, of which he used 36. Zagreb's improved Wiechert seismograph had been installed only about a year before.

His analysis distinguished direct crustal phases from waves refracted through a faster layer below, identifying the extra P- and S-wave phases across a distance range of roughly 300 to 720 kilometres.

Published in 1910, not announced on the day

The listing collapses the earthquake and the discovery into one moment. The earthquake supplied the data; the analysis was published in 1910, and it drew on records from other earthquakes and many European stations rather than a single Zagreb seismogram.

What he demonstrated was a seismic-velocity discontinuity, not a sharp density transition. P- and S-wave speeds depend on elastic moduli as well as density, so no simple proportionality to density applies.

His figure of about 54 kilometres was a model estimate for that regional dataset; modern measurements show Moho depth varying substantially by tectonic setting.

Sources

Researched 24 Aug 2026 6 sources not yet audited

How this was checked

Researched from the web into a fact sheet and rewritten from that sheet. This article has not yet had an independent model audit. How the pipeline works →

What the sources leave uncertain

The date is secure, but numerical parameters are retrospective. A modern study gives Ms 5.78 ± 0.14 and depth 14 ± 7 kilometres, while older catalogues commonly list magnitude 6.0.
The modern epicentre has sizeable uncertainty because many 1909 observatory clocks were poorly controlled; the relocation is 45.57 ± 0.06°N, 16.01 ± 0.08°E.
Published clock times conflict: the modern reanalysis gives 08:59:11.95 for the origin, while some older English accounts say 09:59 Greenwich time. The fact sheet therefore does not attach an exact time.
The approximately 54-kilometre discontinuity depth was Mohorovičić’s model estimate for this regional dataset, not a universal Moho depth.