The Matsu Islands, an archipelago of 36 islands nestled in the East China Sea near the southeastern coast of mainland China, have been administered by Taiwan since 1949. Alongside the Kinmen and Penghu Islands, these outlying territories in the Taiwan Strait rely heavily on vital undersea connections to maintain contact with Taiwan’s main island. Currently linked by two submarine cables—Taima No.2 and Taima No.3—with Taima No.4 soon to join them, the Matsu Islands highlight a broader global vulnerability. The Taiwan Strait is a relatively shallow body of water, averaging just 70 meters in depth, which leaves the heavy fiber-optic lines resting on its seafloor deeply exposed to ship-based hazards like heavy anchors and commercial fishing equipment.
These particular cables have endured an extraordinary run of bad luck. Between 2018 and 2022, the submarine infrastructure connecting Taiwan to the Matsu Islands was severed 27 times, followed by four major outages between 2023 and 2025. This localized crisis mirrors a wider pattern across shallow seas worldwide. In the Baltic Sea, another shallow body of water housing roughly a dozen communications cables alongside energy pipelines, systems have suffered 11 distinct disruptions since 2023.
Globally, an estimated 1.5 million kilometers of subsea communications cables power 607 in-service systems that form the backbone of international telecommunications. If these networks were to fail catastrophically, existing satellite architectures could not possibly absorb the traffic. Entire communities would face total isolation, triggering severe logistical and economic ramifications across societies that have come to take instantaneous global connectivity for granted.
Experts regard these subsea pathways as critical global infrastructure, demanding rigorous engineering to ensure maximum resilience. Yet, modern geopolitical friction and deliberate disruption add an ominous layer of risk to the standard geological and accidental hazards that threaten undersea lines.
Historical precedent underscores how vulnerable these assets have always been to state-sponsored sabotage. On the night of August 4, 1914, just days after Britain declared war against Germany, the British cable ship HMTS Alert sailed from Dover into the English Channel. Trailing grappling hooks along the seabed, the crew hauled up five German transatlantic and European telegraph cables and severed them. The action instantly cut Germany’s direct, secure communications with the Americas, Africa, Spain, and the rest of Europe, forcing Berlin to rely on wireless radio transmissions that British intelligence could easily intercept and decode. A quarter-century later, in September 1939, the British repeated this strategy by cutting German cables running to Lisbon and New York.
Protecting the Cable
Modern subsea cables are thin, polyester-coated conduits resting quietly on the ocean floor, leaving them exposed to a variety of physical threats. To defend against these risks, engineers wrap the inner core of fiber optics and copper conductor sheaths in heavy layers of steel cable jacketing. This extra armor is deployed primarily to mitigate damage from commercial shipping activity.
However, cross-sectional diagrams often present a misleading picture, as they typically depict the heavily armored segments used near the surface and approaching cable landing stations. In contrast, deep-sea segments utilize only the essential inner components: optical fibers and a single thin steel thread wrapped in silicone gel, surrounded by a copper sheath used to power repeaters, and encased in polyethylene. Lacking external armor, these deep-sea sections measure roughly 20 millimeters in diameter, making them about the size of a standard garden hose.
These lines eventually surface at cable landing stations, which are typically secure, unassuming facilities with reliable power supplies. Operators prefer coastal locations featuring minimal maritime or fishing traffic alongside a steep drop-off to the deep ocean floor. Because cables face their greatest threats in shallow coastal waters, minimizing the distance between the deep-sea floor and the dry-land station is a priority. Many jurisdictions also designate official Cable Protection Zones, warning vessel operators to avoid anchoring or commercial fishing in areas where subsea infrastructure crosses the seafloor.
Cable Routing
Designing a subsea route involves balancing multiple competing objectives, though optimizing every factor simultaneously is impossible. Planners must keep routes as short as possible to curb capital expenses and reduce end-to-end signal latency. At the same time, routes should steer clear of active tectonic boundaries and mid-plate rifts. While laying cables deeper offers better protection from shipping traffic and hostile interference, extreme depths make locating and raising lines for maintenance vastly more complicated.
Geographic choke points present another complex challenge. The Bab al-Mandab Strait, for instance, serves as a vital corridor not only for surface vessels moving between the Indian Ocean and the Red Sea but also for 16 major telecommunications cables. In September 2025, several cables were severed within the strait, knocking out an estimated 25% of the total capacity bridging Asia, Europe, and the Middle East, while Houthi-controlled waters remained effectively ungovernable for cable protection operations.
Similar concentration risks exist in the Singapore Strait, a shallow channel handling significant surface traffic and 35 separate cable systems. Further north, the Luzon Strait between Taiwan and the Philippines hosts 18 cables. In December 2006, an earthquake triggered submarine landslides in the strait that severed every cable running through the corridor and buried many beneath deep sediment layers, requiring 11 repair vessels 49 days to restore full operations.
The Black Sea presents acute geopolitical risks and regulatory hurdles. While maritime mines pose a confirmed danger, no direct cable faults have been officially documented from the war in Ukraine, though ongoing conflict makes the region largely inaccessible for routine maintenance. Geopolitics also shapes planning, as operators routinely bypass unstable territorial waters when viable alternatives exist. Furthermore, landing a cable in the United States subjects the project to Federal Communications Commission constraints, including bans on equipment from designated security threats and restrictions against connecting to territories viewed as foreign adversaries.
While land routes are generally avoided, exceptional circumstances occasionally favor them. Connectivity between Asia and Europe typically runs through the Bab al-Mandab Strait, landing in Egypt near the Gulf of Suez before crossing overland to the Mediterranean. An alternative path transits the Persian Gulf, Iraq, and Türkiye, though it requires navigating the Strait of Hormuz or crossing overland through Oman and the United Arab Emirates.
Cable Faults
Industry figures show that roughly 200 submarine cable faults occur worldwide each year, a rate that the International Cable Protection Committee notes has remained relatively steady even as the total global cable inventory expands. The cables themselves demonstrate remarkable durability; rather, human maritime activity above them drives the vast majority of problems. Commercial fishing and vessel anchoring account for an estimated 86% of all faults, with fishing operations causing roughly twice as many incidents as dropped anchors.
Geological events, such as underwater landslides and seabed shifts, account for about 7% of faults, while cable abrasion and component failures in signal repeaters each trigger roughly 4%. Because faults correlate heavily with maritime operations, nearly half of all incidents occur within sovereign territorial waters, with most of the remainder happening inside exclusive economic zones. Only 2% of faults occur on the high seas. Cables remain most vulnerable close to shore in shallow waters where fishing is concentrated, ships drop anchor, and multiple systems converge on coastal landing stations.
National jurisdictions extend far beyond the shoreline. Under the 1982 United Nations Convention on the Law of the Sea, sovereign territorial waters stretch 12 nautical miles from a coast, while coastal states hold exclusive economic rights over seabed resources up to 200 nautical miles—or further under specific continental margin rules. Consequently, any physical intervention, construction, or repair work on a subsea cable within these zones requires formal authorization from the governing state.
Location and Repair
Pinpointing a cable break is remarkably straightforward. An optical pulse sent down the fiber reflects off the break point, providing operators with a precise distance measurement. Once the location is identified, however, the challenge shifts to the marine domain.
Operators must dispatch a specialized repair vessel, which must first secure government permits to operate within the relevant territorial jurisdiction. Upon arrival, the crew drags a grapnel along the seabed to snare the cable, cuts it, brings one end to the surface and secures it to a buoy, and then retrieves the remaining end. Technicians splice a new segment into the line inside an onboard jointing cabin, test the fiber fusion, seal the joint against crushing underwater pressure, and lower the repaired cable back to the seabed with sufficient slack.
Repair costs vary widely based on location, water depth, contractual maintenance arrangements, prevailing weather conditions, and waiting times for an available vessel, though a single repair typically ranges between $500,000 and $1 million. Although the global cable count has climbed, obtaining government repair permits in territorial waters has grown increasingly sluggish. Industry data indicates that average repair response times have more than doubled over the past decade, with the longest single repair in 2023 dragging on for 947 days, largely due to administrative delays in securing clearance.
The physical fleet available for these operations is also under strain. The industry relies on roughly 60 dedicated repair ships operating under regional maintenance agreements that pool operational costs among cable owners. These zone agreements cover the Atlantic Ocean, the North American Western Coastal region, the Mediterranean, Southeast Asia, the Indian Ocean, and the North Pacific out of Yokohama.
Recent analysis by TeleGeography and Infra-Analytics indicates that roughly two-thirds of active cable maintenance vessels will reach the end of their service lives by 2040. Over the same timeframe, total cable length is projected to expand by 48%. Closing that looming infrastructure gap would require an estimated $3 billion investment to build 15 replacement vessels alongside five additional ships merely to sustain current service levels as the network grows.
The immediate dilemma is that individual cable owners have little incentive to foot the bill. Maintenance is traditionally funded through lean, shared agreements optimized decades ago to minimize costs—a model that worked well when the fleet was modern but proves entirely inadequate for financing fleet renewal.
While adding new cable routes remains relatively straightforward—driven largely by commercial incentives from hyperscale cloud providers funding their own construction—no single cable is critical enough to justify the standalone expense of operating a dedicated repair ship. Instead, the industry relies on pooled resources that suffer from chronic underinvestment.
Investigating Cable Outages
Recent high-profile cable outages have drawn intense international scrutiny. Incidents involving merchant ships and fishing vessels in European and Asian waters have frequently raised suspicions of intentional sabotage, though formal investigations often yield complex findings. Swedish authorities found no conclusive evidence of wrongdoing in the case of the Yi Peng 3, ruling the January 2025 damage to the Latvia-to-Gotland cable accidental. Similarly, a Finnish court dismissed charges against the captain of the Eagle S in October 2025, and investigators from the United States and Europe concluded that multiple Baltic incidents stemmed from accidents.
Conversely, legal accountability has been established in other regions. A Taiwanese court convicted the captain of the Hong Tai 58, sentencing him to three years in prison over damage to the Taiwan-to-Penghu No. 3 cable, while the captain of the Newnew Polar Bear faced criminal damage charges in Hong Kong. An April 2026 fault affecting the Matsu Islands was ultimately traced to the wreck of a stranded fishing boat that shifted onto the line during severe weather.
Because anchor dragging occurs frequently among poorly maintained merchant vessels facing rough seas, investigators frequently arrive at inconclusive findings regarding intent, even as ordinary faults accumulate year after year.
Who Is Responsible for Cable Resilience?
The operational architecture of submarine networks has transformed dramatically over recent decades. Traditional telecommunications systems treated each cable as an independent, self-contained entity, making individual owners wholly responsible for their own resilience using ring topologies designed to reroute traffic automatically around localized failures. However, this design required maintaining duplicate capacity, effectively doubling infrastructure costs and reflecting a 50% carriage efficiency.
Seeking cheaper services, Internet service providers began purchasing unbacked capacity across multiple separate cable systems at much higher efficiency rates. When a single cable failed, the ISP absorbed the outage by dynamically rerouting traffic across unaffected networks through IP routing protocols. Consequently, the burden of resilience shifted from individual cable operators to network-level operators. While cable breaks still caused temporary dips in available capacity, the elasticity of modern internet traffic meant that users experienced minor performance degradation rather than catastrophic blackouts.
Resilience and Critical Infrastructure Considerations
Submarine cable systems are constructed by private consortia, maintained through commercial pooling agreements, regulated individually by coastal states, and governed globally by international conventions drafted long before the modern internet existed.
Streamlining repair permits so that operations are not stalled for months awaiting bureaucratic clearance remains a primary industry goal, alongside establishing pre-arranged emergency protocols. Yet, upgrading the aging global repair fleet presents a persistent funding challenge. Without coordinated solutions, the industry risks regressing into a fragmented model where operators must fend for themselves, driving up costs and slowing response times.
Geographic choke points and geological hazards will persist, while extreme weather and shifting ocean sediments continue to threaten subsea lines. Although route diversity provides an effective defense against these disruptions, economic realities often work against it. Cable operators naturally favor the shortest, cheapest paths and lack specific financial motivation to detour around existing infrastructure.
Looking Forward
The telecommunications consortium model that financed and operated nearly all subsea cables thirty years ago has largely vanished, alongside the physical infrastructure and vessels it supported. Approximately a quarter of the world’s current subsea cable length is projected for retirement by 2030, amounting to 845,000 kilometers of fiber-optic infrastructure laid during the massive internet boom of the late 1990s and early 2000s. Historic systems such as TAT-8 and TAT-14 in the Atlantic, SEA-ME-WE-3 linking Europe and Asia, and early trans-Pacific links between Japan, China, and the United States have already ceased operations. Furthermore, more than half of the existing repair vessel fleet will reach retirement age by 2040.
In their place, new cable projects are increasingly funded or controlled by major digital technology firms. This shift raises sovereign risk concerns among nations wary of foreign entities controlling their primary communications conduits, while also complicating traditional maintenance pooling arrangements. By 2028, Google is projected to own the largest share of subsea cables by distance across the Pacific, Atlantic, and Indian Oceans, controlling over half of the total cable inventory in the Indo-Pacific region alone. Similar centralization trends are reshaping the Atlantic basin, granting a single private corporation unprecedented influence over global maintenance networks and repair vessel operations.
This consolidation presents profound challenges for society’s reliance on stable, private, and efficient communications infrastructure. While shifting public telecommunications into private hands during the late 20th century successfully harnessed market capital to build out high-speed packet-switching networks, it also concentrated global connectivity into the hands of a few entities accountable primarily to institutional investors rather than the public interest.
Outside the United States, few economies possess the resources to address these structural vulnerabilities independently. While the European Union has taken steps to fund regional repair vessels and enhance subsea surveillance for security purposes, global resilience ultimately depends on ocean-level infrastructure management. Protecting the digital economy requires coordinated international action to safeguard diverse cable routes, modernize repair fleets, and establish robust governance for the vulnerable seabed foundations upon which global connectivity rests.
