Climate-change-triggered landslide caused the Earth to vibrate for 9 days

A mysterious, globally observed and unprecedented 9-day-long seismic signal in September 2023 was caused by a massive landslide in Greenland. 25 million m3 of rock and ice fell into the remote Dickson Fjord and, in turn, caused a 200 metre-high mega-tsunami that continued sloshing back and forth - a phenomenon called a seiche - in the narrow fjord for 9 days. That is the conclusion of research published in the journal Science.  This movement of a large mass of water generated vibrations through the Earth, shaking the planet and radiating globally observed seismic waves. Never before have scientists observed such an unusual mechanism causing a global seismic signal.

A mysterious seismic signal lasting 9 days was discovered by puzzled seismologists on highly sensitive sensors all over the globe, from the Arctic to Antarctica, in September 2023. The signal looked completely different to frequency-rich earthquake recordings - it contained only a single vibration frequency, like a monotonous-sounding hum.  At the same time, news of a large tsunami in a remote North East Greenland fjord reached authorities and researchers working in the area. The two teams joined forces in a multidisciplinary group involving a unique collaboration of 68 scientists from 40 institutions in 15 countries, combining seismometer and infrasound data, unique field measurements, on-the-ground and satellite imagery, and simulations of tsunami waves. The team also used imagery captured by the Danish military who sailed into the fjord just days after the event to capture the collapsed mountain-face and glacier front along with the dramatic scars left by the tsunami. It was this unique harmony of local field data and remote, global-scale observations that allowed the team to solve the puzzle and reconstruct the extraordinary cascading sequence of events in September 2023. The results are now published in Science.

Seismometers are sensitive scientific instruments that record vibrations traveling through the ground - called seismic waves. Traditionally, seismology focuses on measuring seismic vibrations arising from earthquakes in the ground. However, seismic records can also contain information about movements of large masses on Earth’s surface, such as landslides and water waves. The study found that the landslide was from the collapse of a mountaintop that previously towered 1.2 km above the fjord. The volume of material that collapsed was massive - more than 25 million cubic metres - enough to fill 10,000 Olympic-sized swimming pools. This collapse was caused by glacial thinning at the base of the mountain over recent decades, ultimately caused by climate change.

Numerical simulations, data from local oceanographic sensor network, satellite and on-the-ground imagery confirm that the resulting mega-tsunami is one of the highest seen in recent history. Further out of the fjord, 4 m high tsunami waves damaged a research base at Ella Ø (island) 70 km away and destroyed cultural and archaeological heritage sites across the fjord system. The fjord is on a route commonly used by tourist cruise ships visiting the Greenland fjords. Fortunately, no cruise ships were close to Dickson Fjord on the day of the landslide and tsunami, but if they had been, the consequences of a tsunami wave of that magnitude could have been devastating.

Lead author, Kristian Svennevig, from the Geological Survey of Denmark and Greenland (GEUS) says: “When we set out on this scientific adventure, everybody was puzzled and no one had the faintest idea what caused this signal. All we knew was that it was somehow associated with the landslide. We only managed to solve this enigma through a huge interdisciplinary and international effort”. Kristian Svennevig adds: “As a landslide scientist, an additional interesting aspect of this study is that this is the first-ever landslide and tsunami observed from eastern Greenland, showing how climate change already has major impacts there”.

The signal was so puzzling, that one of the scientific team members tried to recreate the long-lived sloshing effect in their bathtub at home. They failed to simulate the same effect, so it was left to detailed mathematical models to show that the landslide direction, together with the uniquely narrow and bendy fjord channel, was the last missing piece of the puzzle of how climate change rang the Earth for 9 days. The predictions showed that water sloshed back and forth every 90 seconds, the same oscillation period observed in the seismic waves. This perfect match shows how the force of the moving water body, generating a distinct oscillation due to the width and depth of the fjord, creates seismic energy in the crust.

Co-author Stephen Hicks, from University College London says: “When I first saw the seismic signal, I was completely baffled. Even though we know seismometers can record a variety of sources happening on Earth’s surface, never before has such a long-lasting, globally-travelling seismic wave, containing only a single frequency of oscillation, been recorded. This inspired me to co-lead a large team of scientists to figure out the puzzle.” Stephen Hicks continues: “Our study of this event amazingly highlights the intricate interconnections between climate change in the atmosphere, destabilisation of glacier ice in the cryosphere, movements of water bodies in the hydrosphere, and Earth’s solid crust in the lithosphere. Never before anywhere on Earth have we directly recorded water sloshing over days caused by a single event lasting just minutes - in this case a landslide”.

Co-author Thomas Forbriger from Karlsruhe Institute of Technology says: “We wouldn’t have discovered or been able to analyse this amazing event without networks of high-fidelity broadband seismic stations around the world, which are the only sensors that can truly capture such a unique signal”. Thomas Forbriger continues: “Solving a puzzle, presented to us by nature itself, is fundamental research in its purest sense.”

Co-author Anne Mangeney from Université Paris Cité, Institut de Physique du Globe de Paris says: “This unique tsunami challenged the classical numerical models that we previously used to simulate just a few hours of tsunami propagation.  We had to go to an unprecedentedly high numerical resolution to capture this long-duration event in Greenland. This opens up new avenues in the development of numerical methods for tsunami modelling.”

Co-author Thomas Lecocq from the Royal Observatory of Belgium says: “It’s amazing that what started as a routine check of a Belgian gravity sensor turned into a global, multi-disciplinary collaboration, with virtual, online exchanges over 24/7, covering many time zones. In total, more than 8,000 messages were exchanged. Summing up their lengths, it’s over 1 million characters typed in, or… a fascinating 900-page detective novel! I’m happy we proved the source of the vibrations was water sloshing, although I’m a bit sad that some of the imaginative candidates brought up during our “no no’s” idea exchange, such as the “alien slow party” or “new volcano”  were acquitted. We are also happy this adventure led to new collaborations with colleagues all over the world, some of whom are now installing seismic instruments in Dickson Fjord to further advance our understanding of this unique area”.

Co-author Søren Rysgaard from Aarhus University says: “As part of a project where we are establishing a network of real-time under-ice oceanographic stations in Northeast Greenland. We sailed into Dickson Fjord last year to install instruments right in front of the mountain just weeks before it collapsed. It’s eerie to think that the entire fjord was flushed by such a giant wave shortly after. Luckily our instrument survived and we could follow the events unravel in real-time. It is great that our imagery and observation network could contribute to the ground-truthing of the tsunami. Now, we're returning to the area to check the status of our instruments and install new ones— in case there’s another event in the fjord.”

The study concludes that with rapidly accelerating climate change, it will become more important than ever to characterise and monitor regions previously considered stable and provide early warning of these massive landslide and tsunami events.