A common assumption about the modern internet is that international data travels through satellites, beamed up into orbit and back down on the other side of the world. The reality is considerably more physical and, in some ways, more remarkable: the overwhelming majority of international internet traffic travels through fiber optic cables lying directly on the ocean floor, some of them stretching continuously for many thousands of kilometers across entire ocean basins.
These submarine cable systems represent one of the largest and least visible pieces of infrastructure supporting modern life, and understanding how they are actually built, deployed, maintained, and repaired reveals a great deal about why internet connectivity behaves the way it does, why certain regions experience outages when others do not, and why this century-old approach to intercontinental communication has proven remarkably difficult to replace.
What a Submarine Cable Physically Contains
At the center of a modern submarine communications cable sits a bundle of optical fibers, each individual fiber a strand of extremely pure glass roughly the thickness of a human hair, capable of carrying enormous quantities of data as pulses of light traveling along its length at close to the speed of light through glass.
Surrounding this fiber core are multiple protective layers designed to shield the fragile glass from the genuinely hostile conditions of the deep ocean environment, including a waterproof barrier, steel wire strands providing tensile strength, copper conductors carrying electrical power, and heavy outer armoring in sections where the cable faces particular physical risk.
The complete assembled cable is considerably thinner than most people expect for infrastructure of such importance, with deep-ocean sections often no thicker than a garden hose, though shallow-water and near-shore sections carry substantially heavier armoring and can reach the diameter of a human arm to withstand anchors, fishing gear, and shifting seabed conditions.
Why Optical Fiber Carries So Much Data
Optical fiber transmits information as pulses of light rather than as electrical signals through metal, an approach that permits vastly higher data rates because light can be modulated at extraordinarily high frequencies and because glass fiber suffers far less signal loss over distance than copper wire does.
A single fiber can carry many separate data streams simultaneously by using different wavelengths of light, a technique that allows multiple independent channels to travel through the same physical strand without interfering with one another, dramatically multiplying the total capacity available from each individual fiber.
Because a single cable contains multiple fiber pairs and each pair carries many wavelengths, the total capacity of a modern submarine cable system is genuinely enormous, sufficient to carry a very substantial share of all traffic between two continents through a single cable roughly the thickness of a domestic water hose.
Why Signals Need Amplification Along the Way
Even in extremely pure optical fiber, light signals gradually weaken as they travel, meaning a signal sent across an ocean without any intervention would become too faint to detect reliably long before reaching the far shore, a fundamental physical limitation that any transoceanic system has to address.
Submarine cable systems solve this by incorporating repeaters, specialized amplifier units spliced into the cable at regular intervals along its length, typically spaced somewhere in the range of many tens of kilometers apart depending on the specific system design and the fiber technology used.
These repeaters require electrical power to operate, which is supplied through copper conductors within the cable itself, fed from powerful supply equipment at the shore stations on each end, meaning a submarine cable is simultaneously carrying both optical data and a substantial electrical current along its entire length.
How Cables Are Actually Laid Across an Ocean
Submarine cables are installed by specialized cable ships, purpose-built vessels carrying enormous coiled lengths of cable in large tanks below deck, capable of carrying and continuously paying out thousands of kilometers of cable over the course of a single extended deployment voyage.
The laying process requires careful control of the rate at which cable leaves the ship relative to the ship's speed and the depth of water below, since paying out too little cable leaves it suspended under tension between high points on the seabed, while paying out too much wastes expensive cable and creates loops that risk snagging.
Route surveys conducted well in advance of installation map the seabed in detail to identify the safest path, avoiding known geological hazards, areas of heavy fishing activity, existing cables and pipelines, and regions prone to underwater landslides or seismic activity that could damage the cable after installation.
Why Shallow Water Sections Get Special Treatment
The great majority of submarine cable damage occurs in relatively shallow water near coastlines rather than in the deep ocean, because shallow areas see far more human activity including commercial fishing with bottom-contact gear, ship anchoring, dredging, and construction, all of which pose genuine physical risk to a cable lying on the seabed.
For this reason, cables in shallow water are typically buried into the seabed rather than simply laid on top of it, using specialized plows or underwater vehicles that cut a trench, guide the cable into it, and allow the surrounding sediment to settle back over the cable, providing substantial physical protection.
Deep ocean sections, by contrast, are generally laid directly on the seabed surface without burial, since the depth itself provides effective protection from nearly all human activity, and the practical difficulty and expense of burying cable at extreme depth would be prohibitive relative to the genuinely low risk involved.
What Actually Causes Cable Faults
Fishing activity and ship anchors together account for a very large share of all submarine cable faults, typically occurring when bottom-trawling gear or a dragging anchor catches a cable in shallow water and either severs it outright or damages it sufficiently to disrupt transmission.
Natural causes account for a meaningful minority of faults, including underwater landslides triggered by earthquakes, abrasion where a cable rests against rock in areas of strong current, and in some documented cases damage from marine life, though the latter is considerably rarer than popular accounts sometimes suggest.
Component failure within the cable system itself, including repeater malfunction, represents a comparatively small share of total faults, reflecting the genuinely high engineering standards applied to equipment that must operate reliably for decades in an environment where physical access for repair is extraordinarily expensive.
How a Broken Cable Gets Located and Repaired
When a fault occurs, engineers at the shore stations can determine its approximate location with considerable precision by sending test signals down the fiber and measuring how the signal reflects back from the break, a technique that can narrow the fault position to within a relatively small stretch of a cable spanning thousands of kilometers.
A specialized cable repair ship then travels to the identified location and uses a grapnel, essentially a heavy hooked device dragged along the seabed, to snag and lift the damaged cable to the surface, sometimes cutting it deliberately and recovering each end separately for cables in deeper water.
The damaged section is cut out and replaced with a new length of cable spliced into place, a delicate operation requiring the individual optical fibers to be aligned and fused with extreme precision, after which the repaired cable is lowered back to the seabed, a full process typically taking a matter of days to weeks depending on weather, ship availability, and location.
Why Redundancy Matters More Than Individual Cable Reliability
Because individual cable faults are essentially inevitable over a system's operational lifetime, the resilience of international connectivity depends far more on having multiple independent cable routes between regions than on any single cable being perfectly reliable, a design principle that shapes how the global network is planned.
Well-connected regions typically have many separate cables following geographically diverse routes, meaning a fault on any single cable simply causes traffic to reroute automatically through alternative paths, often with effects invisible to ordinary users beyond perhaps a modest increase in latency.
Regions served by only one or two cables face genuinely different exposure, where a single fault can cause severe and prolonged disruption to internet connectivity for an entire country, a vulnerability that has prompted substantial investment in additional routes for previously under-served regions.
Why Satellites Never Replaced Undersea Cables
Satellite communication faces a fundamental physical constraint that cables do not: the sheer distance a signal must travel to reach orbit and return introduces latency that cannot be engineered away, since it is governed by the speed of light across a genuinely long path.
Traditional geostationary satellites orbit at very high altitude, introducing a round-trip delay noticeable enough to meaningfully degrade interactive applications, while newer low-earth-orbit constellations substantially reduce this latency by operating much closer to the surface, though at the cost of requiring vastly more satellites to maintain continuous coverage.
Even setting latency aside, the total data capacity available through satellite systems remains far below what submarine cables provide, meaning satellites serve genuinely valuable roles for remote areas, mobile platforms, and backup connectivity, but are not a practical substitute for the bulk capacity that intercontinental cables carry.
Who Actually Owns and Builds These Cables
Submarine cable systems have historically been built and owned by consortia of telecommunications carriers, groups of companies jointly financing a cable and sharing its capacity, an arrangement that spread the very substantial construction cost across multiple parties who each gained guaranteed capacity in return.
A significant shift over the past decade has been the entry of large technology companies as direct cable owners, financing and building cables primarily to carry their own enormous internal traffic between data centers rather than to sell capacity to others as a telecommunications business.
This shift has meaningfully changed the economics and geography of new cable construction, since these companies build routes serving their specific network needs, which has increased capacity on some heavily used routes considerably while raising questions about how infrastructure investment is distributed across less commercially attractive regions.
How Landing Stations Connect Cables to the Wider Network
Where a submarine cable reaches shore, it terminates at a cable landing station, a secure facility housing the power supply equipment that feeds the repeaters, the optical equipment that converts between the submarine system and terrestrial networks, and the connections into the broader domestic internet infrastructure.
These landing stations represent genuine points of concentration in the network, since a single facility may terminate several cables, meaning physical damage or power failure at a landing station can affect multiple cable systems simultaneously even when the cables themselves remain entirely intact.
The specific locations of cable landing stations are influenced by seabed conditions on the approach, availability of suitable terrestrial network connections, regulatory permissions, and increasingly by security considerations, given the strategic importance of the infrastructure they house.
Why Cable Routes Follow the Paths They Do
Cable routing reflects a combination of physical geography, commercial demand, and political considerations, since a route must avoid the most hazardous seabed terrain while connecting locations with sufficient traffic demand to justify the very substantial investment a new cable system represents.
Certain natural chokepoints appear repeatedly in global cable maps, narrow waterways and straits through which many cables pass because geography offers no practical alternative, creating concentrations where a single incident can potentially affect an unusually large number of separate cable systems.
Political considerations increasingly influence routing decisions as well, with some operators deliberately selecting routes that avoid particular jurisdictions or that provide diversity from existing paths, reflecting growing awareness of the strategic dimension of communications infrastructure.
How Cable Capacity Keeps Growing Without New Cables
A frequently underappreciated aspect of submarine cable systems is that their usable capacity has often increased substantially over their operational lifetime without any change to the physical cable, achieved by upgrading the optical equipment at each end to use more sophisticated signal encoding.
These upgrades work because improvements in how data is encoded onto light, and in how faint signals are detected and reconstructed at the receiving end, allow more information to be pushed through the same physical fiber than was possible when the cable was originally installed.
This capacity growth through terminal equipment upgrades has meant that the effective lifetime of a submarine cable investment is considerably longer than the original design capacity would suggest, though there are genuine physical limits to how far this approach can be extended on any given fiber.
What Happens When Multiple Cables Fail at Once
Simultaneous faults on multiple cables serving the same region represent the scenario most likely to cause genuinely severe and widely noticed internet disruption, and such events have occurred when a single underwater landslide or seismic event damages several cables following nearby routes.
In these situations, traffic that would normally spread across many routes gets concentrated onto whatever capacity remains available, which can cause substantial congestion and slowdowns well beyond the regions directly served by the damaged cables themselves.
Recovery from multi-cable events can take considerably longer than a single fault, both because there are a limited number of specialized repair ships available worldwide and because those ships must work through damaged cables sequentially rather than repairing several simultaneously.
Why This Infrastructure Remains So Invisible
Submarine cables attract remarkably little public attention relative to their genuine importance, largely because the system works well enough, often enough, that most people never encounter a disruption significant enough to prompt curiosity about the underlying physical infrastructure.
The redundancy built into well-connected regions actively contributes to this invisibility, since automatic rerouting means that individual cable faults which would be catastrophic in a less resilient design instead pass entirely unnoticed by ordinary users going about their normal online activity.
Understanding this infrastructure matters nonetheless, because decisions about where new cables are built, who finances them, and how well-protected they are shape which regions enjoy fast, reliable, affordable connectivity and which remain comparatively vulnerable to disruption for reasons entirely outside their own control.
There is also a growing policy dimension to this invisibility, since infrastructure that the public never thinks about tends to receive correspondingly little public scrutiny over how it is regulated, protected, and funded, even as it becomes progressively more central to essential services including banking, healthcare, emergency response, and government administration that increasingly assume constant reliable connectivity as a basic operating condition.
Several governments have consequently begun treating submarine cables explicitly as critical national infrastructure, extending to them the kind of formal protection, monitoring, and contingency planning historically reserved for power grids, water systems, and transport networks, a recognition that reflects how thoroughly modern economies have come to depend on a small number of physical connections lying largely unguarded on the ocean floor.
The physical reality underlying international internet connectivity is far more tangible than most everyday experience of the internet would suggest. Hair-thin strands of glass fiber, wrapped in protective layers and laid carefully across ocean floors by specialized ships, carry the overwhelming majority of data moving between continents, powered by electrical current running the full length of the cable to feed the amplifiers spaced at regular intervals along the entire route. The system's resilience comes less from any individual cable being invulnerable, which none are, and more from having enough independent routes that faults reroute automatically without users noticing. That principle explains both why well-connected regions rarely experience visible disruption from the cable faults that occur regularly worldwide, and why regions served by only one or two cables remain genuinely exposed. As demand continues growing and new owners reshape which routes get built, the geography of this largely invisible infrastructure continues to determine, quite directly, how well-connected different parts of the world actually are, and how quickly they recover when something on the ocean floor inevitably goes wrong.
Sources
- Wikipedia β overview of submarine cable history, construction, and operation
- International Cable Protection Committee β industry data on cable faults, protection, and repair practices
- International Telecommunication Union β global standards and statistics on telecommunications infrastructure
- National Oceanic and Atmospheric Administration β seabed mapping and ocean floor conditions relevant to cable routing
- Nature β peer-reviewed research on optical fiber technology and capacity
FAQ
Does international internet traffic travel by satellite or cable?
The overwhelming majority travels through fiber optic cables on the ocean floor, not satellites, because cables offer far greater capacity and much lower latency.
How thick is an undersea internet cable?
Deep-ocean sections are often no thicker than a garden hose, while shallow-water sections carry much heavier armoring and can reach the diameter of a human arm.
What causes most undersea cable damage?
Fishing gear and ship anchors in shallow water account for a very large share of faults, with natural causes like earthquake-triggered landslides accounting for a meaningful minority.
How are broken undersea cables repaired?
Engineers locate the fault by measuring signal reflection, then a repair ship uses a grapnel to lift the cable, cuts out the damaged section, and splices in a new length.
Why didn't satellites replace undersea cables?
Satellites face unavoidable latency from the distance signals must travel to orbit and back, and offer far less total capacity than submarine cables provide.
About the Author
We reference Wikipedia, International Cable Protection Committee, International Telecommunication Union, National Oceanic and Atmospheric Administration, and Nature to explain the background and current understanding of this topic.
Loved This Article?
Share it on WhatsApp β Share it on WhatsApp
Get more guides in your inbox β Subscribe to our newsletter for weekly surprising stories from Egypt, Saudi Arabia, Dubai, and beyond.