Subsea fiber-optic cables carry roughly 95% of intercontinental data traffic, making them critical infrastructure for finance, cloud computing, military communications, scientific research, and ordinary internet use.
Satellites attract more public attention, but they do not carry the bulk of global communications. A single modern fiber pair can transmit data at terabits per second.
A transoceanic cable may contain 12, 16, or 24 fiber pairs, with total design capacity measured in hundreds of terabits per second. The result is a physical network on the seabed that functions as the nervous system of the global internet.
These cables are engineering systems, commercial assets, and geopolitical pressure points at the same time. Their routes, landing points, ownership structures, and repair logistics shape how data moves across borders.
How Subsea Fiber Cables Work
A transoceanic cable is not a thick pipe full of wires. In deep water, it is often only about the diameter of a garden hose. Inside that narrow structure are glass fibers, strength members, insulation layers, copper conductors, and protective materials arranged to survive decades under pressure, saltwater, abrasion, and tension.
At the center are optical fibers made from ultra-pure glass. Data travels as pulses of laser light through these fibers. Each fiber pair normally uses one fiber for transmit and one for receive, although newer systems can use more complex transmission arrangements.
Dense wavelength-division multiplexing allows many optical channels, each using a different wavelength, to travel through the same fiber.
Over thousands of kilometers, light attenuates. To solve this, engineers install repeaters along the cable, usually every 60 to 100 kilometers.
These repeaters use erbium-doped fiber amplifiers powered by direct current supplied from landing stations at one or both ends of the cable. The copper conductor in the cable carries that power.
A typical modern repeater must operate reliably for 25 years without maintenance. Once it is on the seabed in the middle of the Pacific, sending a technician is not an option.
Component qualification is severe. Devices are pressure-tested, thermally cycled, and aged before deployment. Failures are rare, but when they occur, the consequences can be expensive and politically sensitive.
Cable Anatomy
Cable design changes with water depth and seabed risk.
In deep ocean, where fishing gear and anchors are less likely to reach, cable protection can be relatively light.
A common deep water cable includes:
• Optical fiber pairs
• A stainless steel or metal tube around the fibers
• A copper conductor for repeater power
• Polyethylene insulation
• Steel strength members
• Outer polyethylene sheath
Near shore, the cable becomes much tougher. In shallow water, it may be single-armored or double-armored with steel wires to protect against trawlers, anchors, currents, and rocky bottoms. Shore-end cable can be several times heavier than deep-sea cable.
The transition from deep water cable to armored shore-end cable is planned carefully. Engineers use seabed surveys, burial assessments, and marine traffic data to decide where additional protection is needed.
Route Survey and Installation
A cable route is not drawn as a straight line on a map and then deployed. The path is surveyed in detail before installation.
Survey vessels use multibeam echo sounders, side-scan sonar, sub-bottom profilers, magnetometers, and remotely operated vehicles. The goal is to identify hazards such as steep slopes, volcanic areas, fault zones, coral, wrecks, unexploded ordnance, subsea pipelines, and unstable sediment.
For shallow sections, burial is preferred. A sea plow pulled behind a cable ship cuts a trench and places the cable beneath the seabed, often one to three meters deep depending on soil conditions and risk. Jetting tools may be used in softer sediment.
Burial reduces damage from bottom trawling and ship anchors, which remain among the most common causes of cable faults.
In deep water, the cable is usually laid directly on the seabed. Cable ships control tension carefully so the cable follows the planned route without suspending across underwater ridges or piling into loops.
Slack management matters. Too little slack can strain the cable across seabed features. Too much slack can create loops that snag.
Cable-laying vessels are specialized ships with large turntables that store thousands of kilometers of cable.
They carry repeaters, branching units, plows, remotely operated vehicles, and navigation systems capable of meter-scale precision.
Landing Stations and Terrestrial Integration
A subsea cable becomes useful only after it lands and connects to terrestrial networks.
Landing stations house power feed equipment, optical line terminals, monitoring systems, security controls, and interconnection facilities.
Some landing stations are simple cable termination sites. Others are major hubs tied directly into data centers, internet exchanges, and long-haul fiber routes.
Landing geography has strategic importance. Marseille, Singapore, Mumbai, Djibouti, Alexandria, Fortaleza, Virginia Beach, and Los Angeles have become major cable nodes because of their positions relative to continents, trade routes, and data center markets.
A landing point also creates jurisdictional exposure. The state where a cable lands may regulate operators, demand lawful intercept capability, impose licensing requirements, or restrict ownership.
This makes landing rights as important as seabed route engineering.
Ownership Has Shifted Toward Cloud Giants
For decades, telecom carriers dominated subsea cable ownership. Companies such as AT&T, Verizon, Orange, NTT, Tata Communications, and Telstra invested in consortium cables to serve voice and data traffic.
That model has changed. Google, Meta, Microsoft, and Amazon now finance or co-finance many high-capacity systems because their cloud platforms, content delivery networks, video services, and artificial intelligence workloads require enormous private bandwidth.
Google has invested in systems such as Dunant between the United States and France, Equiano along the west coast of Africa, and Firmina linking the Americas.
Meta has backed 2Africa, a massive system encircling much of Africa with landings in Europe, the Middle East, and South Asia.
Microsoft and Amazon also participate in cable projects and long-term capacity agreements.
This shift changes routing incentives. Traditional carriers built capacity for wholesale telecommunications markets. Cloud companies build cables to connect their own data centers, cloud regions, and edge infrastructure.
The public internet still benefits from added capacity, but private traffic engineering now strongly influences global route design.
Chokepoints and Strategic Risk
Subsea cables follow geography, and geography creates chokepoints.
The Red Sea and the Suez route connect Europe to Asia.
The Strait of Malacca links the Indian Ocean with the South China Sea. Guam serves as a Pacific hub for cables connecting Asia, Australia, and North America.
The waters around Taiwan carry dense cable infrastructure crucial to East Asian communications.
Damage in these areas can have global effects. In 2008, multiple cable cuts in the Mediterranean and Middle East disrupted internet service across parts of Egypt, India, and the Gulf.
In 2024, cable damage in the Red Sea affected systems including AAE-1, Seacom/TGN-Eurasia, EIG, and TGN-Gulf, reducing capacity on Europe-Asia routes and forcing traffic onto alternate paths.
Cable networks are resilient because traffic can be rerouted, but resilience is uneven. Financial centers and cloud regions often have multiple paths.
Smaller island states may depend on one or two cables. If those fail, latency rises sharply or connectivity drops to satellite backup.
Sabotage, Surveillance, and Ambiguity
Subsea cables are vulnerable because they are long, exposed, and difficult to monitor continuously.
Most faults are accidental, caused by fishing activity, anchors, undersea landslides, earthquakes, or equipment failure.
Intentional interference is harder to prove.
A cable can be cut by dragging an anchor, using a grapnel, deploying subsea tools, or interfering with landing infrastructure.
In shallow water, sabotage is technically easier. In deep water, it requires specialized vessels or autonomous systems.
Surveillance is another concern. Fiber-optic cables are difficult to tap without detection, especially compared with older copper systems, but not impossible for advanced state actors.
Landing stations are often more attractive targets than deep-sea cable spans because they concentrate equipment, power systems, and network management interfaces.
The legal environment adds complexity. Under the United Nations Convention on the Law of the Sea, states have obligations related to submarine cable protection, but enforcement is limited.
In exclusive economic zones and international waters, attribution and response remain difficult. A suspicious cable fault may be a crime, an accident, a covert operation, or an insurance event. Evidence can be scarce and slow to collect.
Cable Repair Is Slow, Specialized Work
Repairing a subsea cable requires a cable repair ship, spare cable, trained crews, permits, weather windows, and precise fault localization.
Operators use electrical and optical measurements from landing stations to estimate fault distance. A repair ship sails to the area, retrieves the cable with a grapnel or remotely operated vehicle, cuts and tests sections, inserts new cable, splices fibers, seals joints, and lays the repaired section back on the seabed.
In shallow water, repair can be delayed by permits, military restrictions, fishing activity, and port logistics.
In deep ocean, sea state and distance dominate. A repair may take days if a ship is nearby and conditions are favorable.
It may take weeks if the ship must mobilize from another region.
The repair market is limited. A small number of cable ships cover vast ocean areas under maintenance zone agreements. During regional crises or after natural disasters, multiple faults can compete for the same repair capacity.
Engineering for Higher Capacity
Modern subsea systems push capacity through better fiber, stronger coherent optics, improved repeaters, and smarter spectrum management.
Older systems used fewer fiber pairs and relied heavily on increasing the capacity of each pair. Newer spatial-division multiplexing designs often use more fiber pairs at lower optical power per pair.
This can improve total cable capacity while staying within repeater power limits.
Coherent transmission has transformed subsea performance. Digital signal processing compensates for chromatic dispersion, polarization effects, and nonlinear impairments.
Modulation formats can be adjusted depending on route length and required performance. Shorter systems may support higher-order modulation and greater spectral efficiency. Long transpacific spans require more conservative designs.
Power is a hard constraint. Repeaters draw energy continuously, and power feed equipment has voltage and safety limits.
A cable across the Atlantic differs greatly from one across the Pacific because distance changes optical budget, repeater count, power demand, and achievable capacity.
Security by Redundancy
No single protection method can secure the subsea network. The practical answer is redundancy, route diversity, monitoring, and faster repair.
Countries and operators reduce risk by building geographically diverse cable paths, avoiding excessive dependence on one chokepoint, hardening landing stations, and coordinating with maritime authorities.
Automatic identification system data, satellite imagery, acoustic monitoring, and patrols can help detect suspicious vessel behavior near cable corridors, though none provide perfect coverage.
For island states, redundancy can be expensive but essential. A second cable landing may cost hundreds of millions of dollars, yet dependence on one system can expose a national economy to a single anchor drag or seabed landslide.
Pacific islands, Caribbean states, and remote territories face this calculation directly.
Regulation and National Control
Governments increasingly treat subsea cables as strategic infrastructure. The United States reviews cable projects through processes involving the Federal Communications Commission and national security agencies.
Australia, Japan, the United Kingdom, France, India, and the European Union have also tightened scrutiny around cable ownership, vendors, routes, and landing rights.
Concerns often focus on foreign state influence, espionage risk, supply-chain dependence, and control over repair or maintenance.
Vendor selection can become political. Chinese suppliers such as HMN Technologies have faced restrictions in several projects backed by the United States or its partners. Western suppliers including SubCom, Alcatel Submarine Networks, and NEC remain central to many new builds.
This competition affects developing regions. Countries seeking new connectivity may face pressure over who finances, builds, lands, and operates their cables.
A proposed route can become a diplomatic issue before a cable ship loads its first kilometer of fiber.
The Future of the Seabed Internet
Demand will keep rising as cloud regions expand, artificial intelligence training clusters exchange huge datasets, video traffic grows, and enterprises shift more workloads across borders. More cables will be built across the Atlantic, around Africa, through the Pacific, and into underserved coastal regions.
The engineering trend is clear: more fiber pairs, better coherent optics, improved route planning, and tighter integration with data centers.
The geopolitical trend is just as clear: states will pay closer attention to who owns cables, where they land, who repairs them, and which routes carry critical traffic.
The global internet looks weightless from a browser window. It is not.
It is glass, steel, copper, power, ships, permits, seabed maps, and political risk stretched across the ocean floor.
The next decade of connectivity will be shaped as much by marine engineering and route security as by chips, satellites, and software.
Top comments (0)