Geoff Bennett, director of solutions and technology, and Pierre Mertz, fellow, Infinera
Infinera: Four stages towards a global submarine cable detection network
A 2015 United Nations report estimated that every year, an average of 60,000 people and US$4 billion in assets are exposed to the global tsunami hazard, which can be triggered by certain types of undersea earthquakes or volcanic eruptions.
Seismic vibrations that travel through the earth’s crust move 20 times faster than the tsunami does on the surface – so if scientists can detect the initial earthquake, then they can provide a warning for people to evacuate vulnerable areas.
Most seismic sensors today are located on land, but undersea earthquakes can take place many kilometres from the nearest land-based detector.
Ocean-based measurement buoys – such as DART (Deep-ocean Assessment and Reporting of Tsunamis) buoys – are exposed to extreme weather, are often used as unofficial moorings, and are even vandalised to the point where up to 60% of them may be offline at any one time.
Submarine cables would be a great detection system, and the SMART (Science Monitoring And Reliable Telecommunications) cable initiative is working on enhancing the repeater units that are placed roughly every 80 km along a cable to include seismic, temperature, and pressure sensors.
SMART cables will one day be the gold standard for ocean monitoring, but the first such cable is not expected to be deployed until 2025, and it’s not clear what percentage of new submarine cables will be SMART in the future.
Not surprisingly, many people are looking at ways to enhance existing submarine cables to serve as detectors, and there are many techniques either deployed or in development.
Distributed acoustic sensing (DAS)
This mature, commercially available technique transmits a high-power laser into one of the fibres at the end of the cable. Tiny imperfections in the fibre cause a backward scattering of the signal that can be picked up by a very sensitive receiver. In the receiver, a coherent optical time-domain reflectometer is used to analyse the signal and detect anomalies caused by physical disturbances along the fibre.
DAS can pick up a huge range of effects with extremely high sensitivity, including marine animal sounds, surface ships and submarines, noise from anchors or fishing nets, and, of course, undersea seismic events.
However, DAS requires dedicated hardware that necessitates training for operations personnel and a dedicated fibre that cannot be used for communications, and it only operates over a limited distance – either to the first repeater or up to about 100 km on an unrepeatered cable.
State of polarisation (SOP) measurement
A coherent transponder includes the ability to compensate for changes in the state of polarisation as the signal travels along the optical fibre. These changes can be caused by any kind of stress on the fibre, including temperature changes and movement.
A 2021 paper by Google researchers analysed SOP changes to detect the vibrations from an earthquake near Jamaica using the Curie cable that runs down the west coast of the Americas – a distance of 1500 km. This was an amazing achievement, but SOP on its own is inadequate for the localisation of seismic events. In other words, it could tell something was happening, but not where it was happening.
In May of 2022, a team at the National Physical Laboratory in the UK published a study on a trans-Atlantic cable in which they used a differential technique based on the fact that existing submarine repeaters include a diagnostic loopback component.
This loopback allowed the team to isolate the cause of vibrations to a particular repeater span. The NPL technique uses dedicated optical test equipment, which is not ideal in practical deployments as it’s usually more expensive and requires specialist staff training – especially as cable landing stations have stringent limits on who can gain access.
Most recently, a team from Infinera and Google published a paper at the SubOptic show in February 2023, where a commercial transponder was used as a tone source while still having the ability to carry data traffic, and used a similar loopback technique to isolate seismic events. The technique was deployed on the Curie cable, where it successfully detected an earthquake that occurred on the mainland within a few hundred kilometres of the cable.
Seismic detection with submarine cables is a fascinating area of development. While SMART cables may offer the best quality of data, these cables will not be widely deployed for several years – and possibly a decade for widespread coverage.
Using existing cables is a useful expedient in the short to medium term, and new differential techniques using familiar equipment are showing real promise in what they can detect. While the inter-repeater resolution demonstrated in current trials may not be ideal when used with a single cable, the technique is inexpensive and simple enough to be expanded to a network of cables that would enable signals to be triangulated and significantly enhance detection precision.
