Subsea Cable Monitoring: DAS Covers 600 km of HVDC Cable While Mexico Signs a New Line to Cozumel
On September 28, Sumitomo Electric announced a Distributed Acoustic Sensing (DAS) result on a high-voltage direct current (HVDC) submarine cable exceeding 600 km. The system covered the full length and located a simulated anchor strike within 68.4 m. Two days later, Mexico’s Comisión Federal de Electricidad (CFE) is scheduled to sign the contract for a new submarine power line to Cozumel. Both events belong in the same conversation about subsea cable monitoring. One shows what the technology does today. The other shows where it is still missing.
Three results from the last ten days explain the first half of that sentence. The Cozumel contract explains the second.
What Sumitomo proved on 600 km of HVDC cable
The Sumitomo Electric paper received the Best Paper Award in category B1 PS1 at the CIGRE Paris Session 2026 on August 28. The setup combines the company’s ultra-low-loss fiber and remotely pumped erbium-doped fiber amplifiers with a commercial DAS interrogator from AP Sensing GmbH. As a result, each cable end reaches more than 300 km of sensing range. Both ends together cover a cable exceeding 600 km.
Two figures matter for a power cable operator. First, location uncertainty stays below 200 m, or 0.03 percent of the length. Sumitomo describes that as a factor of 33 improvement over conventional methods. Second, a sea trial off Awaji Island in Japan tested a simulated anchor impact. The system located it with an error of only 68.4 m from the actual point of impact.
Nothing active sits on the seabed, because the amplifiers receive their pump light through the fiber from shore. Therefore, there are no subsea electronics to power, service or recover when something fails. For a buyer, that is the reliability argument, not the range figure.
The economics of subsea cable monitoring: repair ships are the bottleneck
IEEE Spectrum reported on September 26 that DAS already watches the first and last 200 km of each cable. Transoceanic routes, however, run between 6,000 and 20,000 km. The report cites a recent International Telecommunication Union (ITU) study. It counts between 150 and 200 cable accidents per year, with fishing and anchoring as the two main causes. That study comes from the ITU’s International Advisory Body on Submarine Cable Resilience, which approved its final report on July 10, 2026.
The cost side is where subsea cable monitoring earns its budget. According to the same report, a single repair costs roughly 20 times the price of a DAS unit. Moreover, only 20 to 25 repair ships exist worldwide, so a damaged cable can wait weeks for a vessel. A DAS system, on the other hand, can warn an operator up to 15 minutes and 2 km before contact. In other words, the value lies in prevention first and in accurate fault location afterward.
Latin America has a fresh example. On June 24, 2026, an earthquake broke Venezuela’s submarine internet cable 1,800 m off the coast of La Guaira. The repair took weeks. Sensing does not prevent a seismic break. However, fault location is exactly what the Sumitomo trial measured, and a precise location shortens the search for a scarce repair vessel.
Sensing on a cable that carries live traffic
At ECOC 2026 in Málaga (September 20 to 24), fiber sensing was the largest single topic: 60 of 677 papers, according to the MapYourTech conference review. Two field results stand out for subsea cable monitoring.
First, Sikt, ASN and GÉANT ran L-band DAS on a 250 km unrepeatered submarine cable carrying live production traffic. They reported no noticeable deterioration, so the sensing layer needed neither a dark fiber nor a maintenance window. Second, Trinity College Dublin and Tampnet analyzed 122,174 state-of-polarization (SOP) recordings from a deployed North Sea cable. Their detector ranked all five confirmed trawler contacts within the top 13 recordings.
The underlying paper, by Agastya Raj, Alvaro Doval, Tian Tian, Steinar Bjørnstad and Marco Ruffini, describes the method as fully unsupervised. It used no trawl labels in training, no model selection and no hyperparameter tuning. Even so, it surfaced additional contact events that earlier manual reviews had missed. For operators in Latin America, where a labeled history of cable incidents almost never exists, that is the relevant detail.
The same review also lists the limits. SOP monitoring struggles when the event produces a low signal-to-noise ratio, and DAS on a live cable has to share spectrum with the traffic. Neither method replaces a burial survey or a protection zone. Instead, they add a detection layer on top of them. For the basics of landing station DAS, see our earlier article on submarine cable monitoring with fiber sensing.
A new cable to Cozumel, and no monitoring layer on record
CFE issued the award for the Playa del Carmen to Cozumel submarine line on September 15. According to Noticaribe, the contract signing is set for September 30, 2026. The contract is worth 3,563,683,589 pesos and goes to a consortium of Entia de México, Cables Subterráneos and Zhongtian Technology Submarine Cable. Work starts on October 5, 2026, with completion set for February 26, 2028.
Grupo Animal adds the technical outline. The Playacar-Chankanaab II link runs at 115 kV over 33.6 km of line, with a reported value of 184.6 million dollars. The previous cable’s capacity had fallen from 35 MW to 15.5 MW, which is why the project exists. However, neither report mentions a monitoring system for the new cable, whether DAS for external damage, distributed temperature sensing for hot spots or anything else.
That silence is the point of this article. According to the IEEE Spectrum report, standard DAS reaches 200 km from a landing station. A 33.6 km link therefore falls well inside that range, without special fiber or remote amplification. Meanwhile, the regional risk is documented. Panama’s Maritime Authority counts seven submarine systems in its waters and reports losses from cable damage above 11 million dollars between 2008 and 2019. A cable that is being rebuilt because the last one lost capacity is a strong candidate for subsea cable monitoring from day one.
Sector notes: Comcast on land, Luna Innovations under new ownership
On September 29, Comcast announced that it now analyzes vibration on its own fiber to distinguish subway trains from construction equipment near buried lines. The release names no vendor and no kilometer count. Still, a tier-one operator now treats sensing on in-service fiber as routine, not as a pilot.
In the vendor market, Luna Innovations completed its sale to private equity firm TJC LP on September 18, 2026, at 1.39 dollars per share in cash. The reported value is 220 million dollars. The path there included accounting errors and a class action settlement. For a buyer in Latin America, the lesson is that vendor stability and long-term support matter as much as the datasheet.
What to do with this if you own a cable landing
Most readers of this site do not own a 600 km HVDC link. They own a 30 km interconnection, a landing station or the coastal segment of a longer route. For that case, the week’s subsea cable monitoring results translate into three actions.
- Check the fiber you already have. Submarine power and telecom cables usually carry optical fiber. The ECOC results show that sensing can share a fiber with live traffic. So the first question is which fibers exist, where they terminate and whether a port is free at the landing station.
- Ask for anchor and trawl detection, not only fault location. The 15 minute warning window is what avoids the repair cost. However, that window depends on the analytics layer, not only on the interrogator.
- Plan the data side before the hardware. Trinity and Tampnet needed 122,174 recordings to rank five contacts. Decide early who receives an alert, how to check it against vessel traffic data and what counts as a false alarm.
This is the segment where BiiSensing, the first fiber sensing company in Latin America, works: cable monitoring for aerial, underground and submarine power cables. The interrogator stays on shore, with nothing added on the seabed. The usual path is a feasibility assessment of a specific segment, then a proof of concept on fiber that is already installed, and support from site survey through commissioning. If the Cozumel line or a similar project is on your desk, talk to an engineer before the cable ships, while fiber allocation and landing station design can still change.
