Each spring, the European geoscience community gathers in Vienna for the EGU General Assembly, and every year brings new energy, ideas, and collaboration. In April 2025, EGU25 brought together more than 20,000 participants from around the world to share ideas, challenge assumptions, and spark new directions in Earth, planetary, and space-related science.
For the SUBMERSE project, it was a milestone moment: five team members from the consortium presented their work, showcasing how submarine fibre optic cables—traditionally used for telecommunications—are becoming powerful tools for ocean and Earth observation through Distributed Acoustic Sensing (DAS) and other sensing techniques.
Fibre sensing continues in prominence
Just as at last year’s EGU, fibre sensing continues to grow in prominence. This year, several sessions during the first days of the conference focused on DAS and other fibre-optic point and distributed sensing technologies in geoscience.
As one of the most active EU projects in this space, SUBMERSE was present across multiple sessions, with contributions ranging from new machine learning models for submarine DAS data to the introduction of the DAS virtual station concept.
Let’s take a closer look at the researchers who brought SUBMERSE to the spotlight at EGU25.
David Schlaphorst on observing oceanographic variables in DAS data
David Schlaphorst, Postdoctoral researcher in seismology at the Dom Luiz Institute (IDL), University of Lisbon, explored how DAS strain data from a submarine cable near Madeira reveals oceanographic phenomena, such as tides, significant wave heights, and currents.
His analysis showed a near-perfect correlation between DAS signals and conventional buoy measurements, confirming that DAS cable measurements can observe a variety of relevant oceanographic signals such as tides, waves, earthquakes, whales, and more.
“A key takeaway I would say is that it is possible to see the influence of tides on the measurements on the DAS cable really clearly. It matches perfectly well, so we can easily convert one into the other. And I have one image in my talk that basically shows it 1-to-1. You can see where there is high tide, and where it’s low tide. And then on top of that, we can see how the incidence angle of the waves onto the cable changes in this timeframe.”
Additional reading: In his recent blog post for the EGU Seismology ECS division, David shares how DAS on a single submarine cable off the coast of the Portuguese island of Madeira in the Atlantic Ocean was used to detect a variety of signals, including earthquakes, ocean waves and tides, and even whales. Find out more in this article.
Han Xiao on developing the largest submarine DAS dataset
Han Xiao, research scientist in seismic and environmental monitoring at GFZ, presented a pioneering machine learning model – DeepDAS – tailored to detect earthquakes using submarine DAS data. Traditional land-based seismometers and models struggle in marine environments due to unique noise and deployment conditions.
Han’s model, based on the DeepLab v3 neural network architecture, was trained on nearly 92 million seismic records (manually and semi-automatically labeled) from multiple international submarine sites, outperforming previous models by a large margin (F1 score 0.89 vs. 0.53).
His work also produced the first and largest open-source submarine DAS earthquake dataset, with over 6,300 seismic events. This resource is expected to become a benchmark for future submarine DAS research.
“Within the SUBMERSE project, we built the first and also the largest submarine DAS dataset. This is a great benefit from this project.”
Chris Atherton on how SUBMERSE is progressing
Last year at EGU24, Chris Atherton, Senior Research Engagement Manager at GÉANT, presented the project and shared the major discovery of the first year: that SUBMERSE was able to deploy DAS signals in harmonious coexistence with production dense wavelength division multiplexing (DWDM) traffic over a telecom fibre.
This year, and two years into the project, Chris highlighted several new achievements:
- Development of an automated, machine-learning-based algorithm for analysing earthquake waveforms and measuring ocean surface gravity waves
- Successful deployment of state-of-polarisation (SOP) measurements synchronised with an atomic clock in one of the most remote locations in the world
Throughout his presentation and in other conversations, Chris emphasized SUBMERSE’s unique approach of integrating existing research infrastructures, network providers, and industry to enable federated, global monitoring of Earth’s systems beneath the ocean with submarine cables.
“SUBMERSE project is really important because it brings together so many different worlds. We’ve got NRENs who provide services for research and education. We’ve got research infrastructures who do the research and provide education. What SUBMERSE is doing is bringing these two worlds together in a completely different way, so that we’re now able to generate data from the infrastructure that we have. And that gives us the ability to gather insights into the areas on our planet that are really hard to reach.”
Miguel Gonzalez-Herraez on going beyond conventional DAS systems
Miguel Gonzalez-Herraez, Professor of Photonics at Universidad de Alcalá, reviewed the foundational principles of DAS and introduced two innovative systems currently under development in his lab.
One system achieves centimetre-scale gauge lengths over a sensing range of approximately 1 km, and the other offers conventional performance over tens of kilometers while eliminating 1/f instrumental noise, making it ideal for very long-term monitoring of slow oceanographic processes like tsunamis and tidal waves. His work highlights how submarine DAS can address gaps in underwater seismic monitoring where current, more traditional land-based solutions fall short.
“I’m involved in developing DAS technology, particularly the DAS technology that we developed in SUBMERSE is meant to be more stable in the long term. Traditional DAS technology is generally good for measuring short-term events. The DAS technology that we develop in this project is more oriented toward measuring processes that takes hours, and even days. For instance, the DAS technology that we develop is more suitable for tsunami early warning, for detecting infragravity waves, and tidal waves.”
Jan Petter Morten on DAS virtual stations for rapid earthquake location
Jan Petter Morten, Principal Engineer specialized in signal processing at ASN, introduced the concept of DAS virtual stations to allow for a much quicker detection and location of earthquakes. He focused on the integration with existing earthquake monitoring systems and showed that edge computing and streaming from the DAS interrogation of submarine cables enhances rapid earthquake location when processed in conjunction with data from terrestrial seismometer stations.
“My key takeaway is that you can start to utilize DAS today. There are a lot of things that have already been implemented. For example, you can use DAS on telecoms’ fibers with traffic already. By doing just a few simple steps you can plug the DAS data into existing workflows.”
Current milestones and what’s coming next
The talks, sessions, and informal discussions at EGU25 reflected growing interest in fibre sensing, and SUBMERSE was at the centre of that momentum. Beyond the technical results, it’s clear the project is shaping how the wider community thinks about infrastructure, data, and collaboration.
Two years in, significant new developments and milestones have taken place at our sites in Norway, Greece, and Portugal, including:
- DAS data generation across all countries and sites: 2 in Norway, 2 in Greece, and 2 in Portugal
- MiniSEED integration set up in Portugal (with Greece and Norway coming soon), enabling DAS data to be openly and securely available to seismology researchers through archives such as ORFEUS and EIDA
- DAS deployment in Greece: it now allows monitoring of the Kefalonia Transform Fault Zone (KTFZ), providing crucial decision-making information for civil protection
- Software and data architecture enhancements to ensure scientists and researchers can access near real-time DAS data
We believe that SUBMERSE’s work is setting the stage for future innovations in fibre sensing, enabling research in areas previously out of reach. Stay tuned for upcoming posts covering our participation at SubOptic 2025 and TNC25, where we will share more stories and insights from our journey.