Volcanic Crisis Unraveled Using Fiber Optic Sensing of a Sub-haul Cable on the Reykjanes Peninsula, SW Iceland
Description:
We present the first(?) fully operational volcano early-warning system based on distributed acoustic sensing (DAS) data. The system is operated by the Icelandic Meteorological Office (IMO), the State Volcano Observatory, and represents a major advance in real-time monitoring of volcanic activity in Iceland. Iceland is one of the most tectonically active regions on Earth, hosting 32 active volcanoes and forming the only subaerial expression of the Mid-Atlantic Ridge. On the Reykjanes Peninsula in southwest Iceland, volcanic activity resumed in 2021 after ~800 years of quiescence. Since then, 2 volcanic systems have erupted repeatedly: Fagradalsfjall (3 eruptions between 2021–2023) and Sundhnuksgigarod (9 eruptions since December 2023), threatening the town of Grindavík and critical infrastructure, including a geothermal power plant. Basaltic fissure eruptions in the area produce fast-flowing lavas that can inundate roads and infrastructure within less than an hour from eruption onset, making reliable early warning essential for evacuation and hazard mitigation. The warning system builds on a collaboration initiated in early 2024, when a DAS interrogator detected a low-frequency signal, related to minute-scale ground deformation prior to eruptions. This signal has now been observed before 6 eruptions, each time providing a clear and actionable warning. In January 2025, IMO permanently deployed a DAS system monitoring 2 fiber-optic sub-haul cables with a total length of ~150 km. Real-time processing uses a simple amplitude-stacking threshold to trigger automatic alerts in the monitoring center. In addition, data from both fibers are used to produce automatic distributed dike models showing magma-induced deformation. The July 2025 eruption demonstrated excellent performance, correctly predicting both timing and location. The system detects a wide range of volcanic and microseismic signals and remains effective during severe weather when conventional seismometers degrade. To date, the system has successfully warned for all 6 eruptions, with lead times between 40 minutes and 3 hours, and zero false or missed alerts.
Session: Fiber-Optic Sensing Applications in Seismology and Environmental Science - V
Type: Oral
Date: 4/17/2026
Presentation Time: 04:45 PM (local time)
Presenting Author: Kristin Jonsdottir
Student Presenter: No
Invited Presentation: Yes
Poster Number:
Authors
Kristin Jonsdottir Presenting Author Corresponding Author kristinj@vedur.is Icelandic Meteorological Office |
Yesim Cubuk-Sabuncu yesim@vedur.is Icelandic Meteorological Office |
Palmi Erlendsson palmier@vedur.is Icelandic Meteorological Office |
Jiaxuan Li jxli@cougarnet.uh.edu University of Houston |
Vala Hjörleifsdóttir valah@ru.is Reykjavik University |
Ettore Biondi ettore88@sep.stanford.edu Stanford University |
Ingvar Kristinsson ingvar@vedur.is Icelandic Meteorological Office |
Matthew Roberts matthew@vedur.is Icelandic Meteorological Office |
Kristin Vogfjord vogfjord@vedur.is Icelandic Meteorological Office |
Einar B Gestsson ebg@vedur.is Icelandic Meteorological Office, Reykjavík, , Iceland |
Jan P Morten jan_petter.morten@asn.com Alcatel Submarine Networks, Trondheim, , Norway |
Zhan Zhongwen zwzhan@caltech.edu California Institute of Technology, Pasadena, California, United States |
Volcanic Crisis Unraveled Using Fiber Optic Sensing of a Sub-haul Cable on the Reykjanes Peninsula, SW Iceland
Category
Fiber-Optic Sensing Applications in Seismology and Environmental Science