Most people picture the internet as something airy. A signal. A cloud. A screen that responds when tapped. The real internet is far less magical and far more physical. It runs through glass fiber, landing stations, power systems, route agreements, data centers, and maintenance ships. Under the oceans, nearly 600 cable systems and more than 1.5 million kilometers of submarine cable tie continents together. ITU says submarine telecom cables carry more than 99% of international data exchange. TeleGeography’s current map tracks 597 active or under-construction systems and 1,712 landings.
Table of Contents
That physical network does more than keep social media alive. It carries cloud traffic, business software, international payments, government data, software updates, medical records, streaming, wholesale internet transit, and the hidden machine-to-machine chatter that keeps companies running. Research published through the World Bank and IFC also shows that submarine cable capacity is tied to lower internet prices and better access, especially in emerging markets. Cheap, fast international bandwidth is not a luxury layer on top of the economy. It is part of the economy.
So what would happen if undersea internet communication suddenly stopped? Not the movie version. Not a single dramatic blackout where every screen goes dead at once. The first truth is messier. The world would not lose all connectivity in one blow. It would lose the smooth global version of the internet, and the damage would fall unevenly. Countries with dense redundancy, strong terrestrial links, local peering, and domestic hosting would limp along. Countries with only a few international exits could be cut off almost completely. Real outages in West Africa, East Africa, Bangladesh, and Tonga already show the outlines of that pattern.
This matters because cable failures are not rare. OECD says submarine cables suffer roughly 150 incidents each year, with around 40% linked to fishing vessels and ship anchors. ICPC notes that dragged anchors alone account for about 30% of incidents annually. Most faults are not spectacular. They are the result of ordinary physical risk: ships, storms, earthquakes, underwater slides, aging equipment, and route concentration. A single cable fault is normal maintenance territory. A broad interruption across the submarine layer would be a system shock.
The internet is more physical than most people think
The modern internet is often described as decentralized, and that is true in one sense. No single company or government owns the whole thing. Routing is distributed. Networks make local decisions. Data can take many paths. But that should not be confused with weightlessness. The internet is decentralized in governance, yet highly dependent on concrete infrastructure in very specific places. A surprising amount of global connectivity passes through a relatively small set of seabed routes, coastal landing sites, and high-capacity terrestrial corridors.
That is also why submarine cables matter more than satellites for the core of global traffic. A modern cable can carry enormous throughput. TeleGeography’s cable FAQ uses MAREA as a reference point at 224 Tbps. By comparison, satellite systems are useful for reach, mobility, emergency backup, and remote access, but they do not replace the dense carrying power of subsea fiber. Reuters noted during Tonga’s 2022 outage that submarine fiber has far greater capacity than satellites, which is exactly why a cable cut can isolate a country so sharply even when satellite backup exists.
The economic side is just as important as the technical side. World Bank and IFC research argues that more submarine cable capacity lowers international connectivity costs, improves competition, and can cut retail internet prices significantly in some markets. That effect disappears quickly when capacity is scarce or concentrated. A cable break is not only a connectivity problem. It is also a price, competition, and service-quality problem. Places with weak redundancy pay for that weakness twice: once in normal conditions through higher costs, and again during outages through harsher disruption.
That helps explain why the same physical event produces radically different outcomes from one country to another. A cable-rich region with several landings, cross-border backhaul, local Internet Exchange Points, cloud regions, and content caches can absorb traffic shifts and keep a lot of digital life moving. A cable-poor country with little local interconnection and heavy dependence on foreign services is far more exposed. The map of cables matters, but the map of dependency matters more.
The first break would be unevenness
If submarine internet communication stopped suddenly, the first public experience would be confusion. Some services would load. Others would stall halfway. A messaging app might deliver plain text but fail to sync media. A domestic news site might open because it is cached locally, while the login button fails because its identity service sits in another country. A bank app might show yesterday’s balance but fail on a real-time transfer. The opening phase would feel less like darkness and more like digital unreliability spreading by the minute.
That pattern showed up in the African cable outages of 2024. In West Africa, multiple submarine cable faults disrupted connectivity across 13 countries, but not with equal severity. In East Africa, damage to SEACOM and EASSy sharply reduced connectivity in some places while others kept partial service through workarounds and rerouting. Reuters and Cloudflare both described rerouting efforts during these incidents, which softened the blow in some markets while adding congestion and slower service. When the main arteries fail, the network tries to save itself. The rescue paths are real, but they are narrower.
That narrower capacity is the key. Traffic does not disappear when a route fails. It looks for another path. When many paths fail at once, the surviving ones fill up fast. Latency rises. Packet loss rises. Systems start timing out. Users often misread that as “the app is broken” or “the website is down.” The deeper cause is that the application is waiting on remote services that are now too slow, too congested, or unreachable. Modern digital services do not fail in one clean piece. They fail in layers.
A second source of confusion would come from cloud dependence. Many companies think of themselves as local because their customers and staff are local. Their tools are often not. Payroll, CRM systems, document storage, identity management, anti-fraud checks, call-center platforms, software deployment, and customer analytics are frequently hosted abroad. A country can keep power, mobile access, and domestic fiber intact and still find large parts of normal business life stalling because the service logic lives elsewhere. A broad undersea outage would expose how much “local” digital activity is really cross-border.
What breaks fastest after a major undersea cable shock
| Service layer | What users would notice | Why it breaks quickly |
|---|---|---|
| Authentication and logins | Sign-ins hang or fail | Identity platforms often sit in remote clouds |
| Payments and card checks | Timeouts, retries, declined transactions | Verification and settlement rely on stable international links |
| Business software | CRM, storage, dashboards, SaaS tools become erratic | Many core workflows depend on offshore platforms |
| Media and communications | Streaming degrades, video calls drop, uploads fail | Backup routes congest and latency rises fast |
This table does not describe a universal sequence that would look identical everywhere. It sketches the most likely failure order in countries that depend heavily on offshore cloud services and international backbones. The common thread is not the app itself. It is the remote dependency behind the app.
Geography would decide the pain
A sudden undersea communication break would not hit the world evenly because the world is not connected evenly. Europe offers one end of the spectrum. Cloudflare’s review of Baltic cable cuts found little to no observable national-level impact in the affected countries because Europe had enough route diversity and redundancy to absorb the faults. That does not mean the event was harmless. It means the regional network was dense enough that users barely felt it. Redundancy is invisible right up to the moment it saves you.
Island states sit at the other end. Tonga remains the clearest example because the mechanism is so easy to see. The January 2022 Hunga Tonga-Hunga Ha’apai eruption damaged the 827-kilometer cable linking Tonga to Fiji and isolated the country’s telecommunications from the rest of the world. The World Bank’s rapid damage estimate describes that isolation plainly. Reuters reported that repairs could take days or weeks and later reported that the repair ship needed 20 days to replace a 92-kilometer section before service was restored. That is what route concentration looks like in real life: one break, one nation, one abrupt collapse in outside connectivity.
West Africa and East Africa sit in the middle of the spectrum, which makes them especially useful examples. The 2024 outages there did not create a uniform continental blackout. They produced a patchwork. Some countries suffered near-total outages. Some saw degraded service. Some carriers rerouted more effectively than others. The lesson is not simply that Africa needs more cables, though more diversity helps. The lesson is that resilience lives above the cable layer too: terrestrial backup routes, IXPs, operator cooperation, and local caches all change the outcome.
Bangladesh offers a more hopeful case. During a submarine cable outage, Internet Society reported that the country managed by leaning on terrestrial links through India and on locally cached content. That is the kind of detail that matters in a real crisis because it turns an absolute outage into a degraded service period. A user may complain about slowness, but slowness and isolation are not the same thing. A nation with fallback paths stays in the conversation. A nation without them watches the conversation continue somewhere else.
Cloud services, payments, and logistics would feel it first
The public would notice the outage through consumer pain, but the deeper damage would run through systems that do not usually sit in public view. Finance is a clear example. ITU’s submarine cable resilience materials describe cables as essential to communication, finance, cloud services, and vital infrastructure. That sounds broad, but the point is concrete. International payments, settlement systems, card authorizations, fraud screening, compliance checks, trade documentation, and treasury operations all lean on stable cross-border connectivity. Money moves through law and software, but it still depends on physical transmission.
A serious undersea interruption would not instantly erase every payment. Domestic transactions would still clear in places with strong local systems. Cash would still exist. Some card networks might keep limited functions alive. The sharper problem would appear at the edges where domestic systems rely on foreign verification, international gateways, offshore fraud scoring, or remote back-office platforms. That is where timeouts and delays would pile up. Businesses would discover very quickly that a payment stack can look national on the front end and remain global underneath.
Cloud services would be hit just as fast. Enterprises now rely on remote storage, SaaS applications, cross-region backups, security services, machine learning APIs, software repositories, and identity providers spread across continents. OECD’s resilience report stresses the growing importance of cloud integration, software-defined networking, and CDNs in modern communications resilience. Those tools help in a crisis because they improve adaptability and rerouting. They also make it easier to forget how international the service chain really is. The cloud does not float above geography. It sits on top of geography and inherits its weak points.
Logistics would be the third major shock area. Ports, airlines, warehouses, customs, freight forwarders, retailers, and manufacturers all depend on a flow of confirmations, manifests, bookings, and tracking data that crosses borders constantly. A factory may still have electricity, staff, and raw material on site, yet fail to dispatch goods because the software chain that confirms shipment slots or customs status is degraded abroad. This is why infrastructure outages often look administrative rather than dramatic. The cable break is at sea. The visible damage shows up in invoices, schedules, bookings, payroll files, and unresolved exceptions.
A domestic internet could survive in fragments
The phrase “internet outage” can be misleading because it suggests a single switch. In reality, many countries have a domestic internet layer that could keep functioning even if international connectivity collapsed. Internet Exchange Points allow local networks to exchange traffic directly instead of sending domestic traffic abroad and back again. Internet Society describes IXPs as shorter, cheaper, and more resilient routes for local traffic. That becomes extremely important during external shocks. A country that keeps local traffic local has already bought itself time.
This is not theoretical. Internet Society’s reporting on African outages repeatedly highlights the role of IXPs, cache-fill programs, and local interconnection. The Foundation’s resilience materials make the same point in broader language: local peering and local content help the internet keep working when long-distance paths are under stress. A domestic news site hosted locally, a banking portal with local interconnection, or a cached copy of popular content can remain accessible even while international services struggle. People often discover the value of local infrastructure only after losing the global layer they assumed would always be there.
Still, “local internet” has hard limits. A site can be hosted domestically while relying on foreign DNS, foreign login systems, foreign analytics, or foreign developer services. A public portal can have local servers and an offshore payment gateway. A hospital can store patient data locally and still depend on remote cloud tools for imaging, procurement, or backup. The modern digital stack is stitched together from many small dependencies, and undersea cable disruption would tug at all of them.
The countries that would cope best are the ones that have already done the unglamorous work: domestic hosting for essential services, local peering, content caches, more than one international exit, and terrestrial cross-border fallback. Internet Society’s Bangladesh case and its West and East Africa reporting point in the same direction. Resilience rarely looks dramatic when it is built well. It shows up as a crisis that feels smaller than it should have.
Satellites would help, but not enough
Whenever submarine cables come up, satellites enter the conversation almost instantly. That instinct makes sense. They are visible, modern, and outside the seabed itself. They are also genuinely useful. Satellite systems can restore a thin line of communication, connect remote communities, support emergency operations, and keep some critical services alive while repairs are under way. ICPC notes that low-level satellites can play an important role in bridging communication gaps when cables are damaged. As emergency support, satellites matter. As a wholesale replacement for the submarine internet, they do not.
The reason is mostly capacity. TeleGeography’s cable FAQ shows how much traffic a modern cable can handle, and Reuters’ Tonga reporting underlined the gap by noting that submarine fiber carries far more data than satellites. That gap is the whole story. Undersea fiber is the heavy-load infrastructure of the modern internet. Satellite networks are better thought of as complementary layers with different strengths: mobility, reach, temporary restoration, and service in places where laying fiber is too costly or too slow. They are lifeboats, not substitute oceans.
There is another catch. Backup only works if the backup is already available. A country or company needs terminals, contracts, permissions, staff familiarity, and emergency procedures in place before the break happens. Otherwise the technology exists in theory and arrives too slowly to matter. Tonga’s later internet troubles, including the strain around cable damage and fallback options, showed how quickly emergency communications become tangled in regulatory and operational questions. Even without those extra complications, no country should plan on satellites carrying normal national broadband life after a major undersea shock.
The sensible architecture is layered. Submarine cables carry the global load. Terrestrial links provide regional escape routes. IXPs and caches keep local traffic local. Satellites cover remote areas and emergency gaps. A system built on only one of those layers is brittle. A system built on several has room to breathe when one of them fails.
Repairs would be slower than the public expects
Most people have a poor feel for cable repair times because they imagine digital problems as software problems. A website breaks, someone patches code, service returns. Submarine cable repair belongs to a different world. A fault has to be located. A specialized repair vessel has to be available. Weather has to cooperate. Spare parts have to be in reach. Access and scheduling issues have to be solved. Then the cable has to be lifted, cut, repaired, tested, and laid back down. This is maritime engineering, not a remote restart.
Real events make that timescale easier to grasp. Reuters reported that Tonga’s main cable outage could leave the country cut off for days or weeks, and later reported that repairs took 20 days for one 92-kilometer damaged section. Internet Society’s East Africa report says the May 2024 damage was fully repaired by 3 June, roughly three weeks later. Those are not outlier numbers from a bygone era. They are recent reminders that undersea infrastructure comes back on ship time, not app time.
Cost adds another layer. ICPC says anchor-related cable repairs typically cost between roughly £500,000 and £1 million per incident. That figure covers the physical repair, not the wider economic loss from congestion, delayed transactions, business interruption, or emergency workarounds. In a large multi-cable failure, the economic harm can easily outrun the repair bill because the surviving routes become overloaded and degraded during the wait. The network pays for the break long before the repair invoice arrives.
Repair speed also depends on preparedness. ITU’s cable resilience work now puts heavy emphasis on faster repair, permit efficiency, maintenance readiness, and better cooperation. That focus is practical. A network is not resilient just because breaks are rare. It is resilient because a break does not turn into prolonged isolation. Fast repair is part of capacity. A cable system with slow restoration is effectively smaller and weaker than it looks on paper.
Real resilience is built before the outage
The most useful answer to a cable-failure question is not fear. It is design. The countries and companies that would handle a sudden undersea disruption best are the ones that treat resilience as everyday infrastructure, not as emergency rhetoric. OECD’s report keeps returning to a few themes: redundancy, diversity, interconnection, cloud-aware network design, and operational readiness. Internet Society’s reports from Africa and Bangladesh add texture to that list by showing what those ideas look like on the ground.
More than one cable matters. More than one landing station matters. Terrestrial fallback matters. Local IXPs matter. Local hosting for state services, financial rails, healthcare systems, and major media matters. Content caches matter. Training matters. Cooperation among operators matters. None of this has the glamour of a launch announcement. All of it decides whether a nation experiences a cable fault as an inconvenience, a slowdown, or a national communications crisis.
There is also a market dimension that should not be ignored. World Bank and IFC research shows that more submarine cable capacity lowers prices and supports broader internet access. That means resilience policy is not only about surviving rare outages. It is also about daily affordability, competition, and service quality. A market with thin capacity and few routes is expensive on an ordinary day and fragile on a bad day. The same investments that improve resilience often improve access and cost in normal conditions.
Cloudflare’s resilience framework is useful here because it shifts attention away from headline events and toward measurable conditions: route diversity, network topology, interconnection depth, and concentration risk. That is the right frame. A country with fewer headlines and better topology may be safer than a country with grand infrastructure claims and hidden dependency on a small number of upstream paths. The strongest internet is not the loudest one. It is the one with options.
The internet would survive, but its shape would change
The cleanest answer to the original question is this: if undersea internet communication suddenly stopped, the world would not lose digital life altogether. It would lose the frictionless, cross-border version of digital life that people now mistake for normal nature. Distance would start feeling like distance again. A file stored on another continent would stop feeling local. A payment crossing borders would stop feeling instant. A call to family abroad would stop feeling trivial.
That is the hidden achievement of submarine cables. They do not merely carry data. They erase remoteness so effectively that remoteness now feels like a bug. A large undersea interruption would reverse that trick. Highly connected regions would still function, though badly bruised. More vulnerable countries would face genuine isolation. Businesses would discover the true geography of their software. Governments would triage bandwidth. Banks would protect core functions. Households would find that some familiar services still worked while others had quietly depended on the other side of the world.
That is why submarine cables deserve more public attention than they usually get. They are not niche telecom assets sitting somewhere below normal life. They are part of the basic machinery that keeps modern societies synchronized. Their value becomes obvious only when they fail, and by then the hard choices have already arrived. The best response is not panic. It is to build more routes, more local capacity, better repair readiness, and a less naive view of how physical the internet really is.
FAQ
Submarine telecom cables carry the overwhelming majority of international data exchange. ITU says they handle more than 99% of international data exchange.
No. The likely outcome is uneven degradation, not one clean global blackout. Domestic networks, cached content, and local services could keep working in many places while international services fail or slow sharply.
Island states, remote markets, and countries with few international exits would suffer the most. Tonga’s 2022 outage is the clearest recent example of how one damaged cable can isolate a country.
Because the surrounding network had strong redundancy. Cloudflare found that recent Baltic cable cuts had little to no observable impact at the country level because Europe had enough alternative paths.
Partial failure. Logins would hang, cards might fail, video calls would drop, and cloud-based apps would become erratic before people understood the common cause.
Not everywhere. A phone network can still function domestically while international services carried over it become slow or unreachable. The exact result would depend on how much of the mobile operator’s backbone and service stack relies on external links.
Yes, if they are hosted locally and if local networks exchange traffic through IXPs rather than sending it abroad and back. That is one of the biggest practical benefits of strong local interconnection.
They let domestic traffic stay inside the country or region. That lowers cost in normal times and improves resilience when international routes are damaged.
Many would. Remote identity providers, storage, SaaS tools, security systems, and cross-region control planes depend on stable international connectivity.
Because many payment flows rely on offshore verification, international gateways, remote fraud checks, and stable transmission between institutions. Cables are part of the infrastructure behind those flows.
Yes. OECD says submarine cables suffer roughly 150 incidents each year. Most are not spectacular, but they are common enough that resilience planning matters.
Fishing activity, dragged anchors, natural hazards, and routine physical wear are major causes. OECD says around 40% are caused by fishing vessels and ship anchors, and ICPC says dragged anchors alone account for about 30% annually.
Because a dragged anchor can cut or damage cable in shallow water and, in some cases, affect more than one cable at once. That raises both repair costs and the risk of major disruption.
Only in a limited way. Satellites are useful for emergency access and remote connectivity, but they do not match the carrying power of subsea fiber for normal large-scale broadband use.
Often weeks, not hours. Tonga’s 2022 repair took 20 days for a major damaged section, and East Africa’s 2024 outage took roughly three weeks to fully repair.
Because cable repair needs specialized ships, good weather, spare parts, fault location work, and careful physical handling at sea. It is a maritime engineering job.
Some domestic banking functions probably would, especially where core systems are hosted locally. Cross-border transfers, external verifications, and cloud-linked features would be much more vulnerable.
Yes. Route diversity, more than one cable landing, terrestrial fallback, IXPs, local hosting, content caching, and faster repair readiness all improve resilience.
The internet holds up best where there are already alternatives in place. The systems that cope well are the ones built with redundancy and local interconnection before anything breaks.
Author:
Jan Bielik
CEO & Founder of Webiano Digital & Marketing Agency
This article is an original analysis supported by the sources cited below
Submarine Cable Resilience
ITU overview of the role submarine cables play in international data exchange, cloud services, finance, and critical connectivity.
Submarine cable resilence
ITU background note summarizing the scale of the submarine cable system and the pressures it faces.
International Advisory Body for Submarine Cable Resilience
ITU page explaining the current global effort to improve repair speed, coordination, and cable resilience.
Enhancing the resilience of communication networks
OECD overview of the report on communication-network resilience, including redundancy, diversity, and network design.
Enhancing the resilience of communication networks (EN)
OECD full report with figures on cable incidents, common causes of faults, and resilience policy.
Submarine Cable FAQs
TeleGeography reference page with current cable length estimates and cable-capacity examples such as MAREA.
Submarine Cable Map 2025
TeleGeography’s current interactive map showing cable systems and landing points around the world.
Will New Satellites End the Dominance of Submarine Cables?
TeleGeography analysis explaining why satellites complement rather than replace undersea fiber.
Damage to Submarine Cables from Dragged Anchors
ICPC paper on one of the most common physical causes of cable faults and the costs of repair.
Submarine Cable Protection and the Environment
ICPC publication covering environmental risk, resilience, and the role of satellite backup in cable emergencies.
2024 West Africa Submarine Cable Outage Report
Internet Society report on the March 2024 outages affecting multiple countries on the West African coast.
2024 East Africa Submarine Cable Outage Report
Internet Society report on the May 2024 East Africa outages and the repair timeline.
Policy Brief: Enhancing the Resilience of Submarine Internet Infrastructure
Internet Society policy brief focused on resilience, route diversity, and practical ways to reduce outage impact.
Internet Exchange Points (IXPs)
Internet Society explainer on local interconnection and why IXPs improve cost, speed, and resilience.
What Can We Learn From Africa’s Multiple Submarine Cable Outages?
Internet Society Pulse article connecting the African outages to local peering, cache fills, and resilience work.
Bangladesh Coping With Submarine Cable Outage Thanks to Indian Terrestrial Cables, Local Content Caches
Internet Society case study showing how terrestrial backup and local caching softened a cable outage.
Ever wondered how the Internet stays on? It’s all about Resilience!
Internet Society Foundation article on local peering, resilience, and the infrastructure behind stable connectivity.
Undersea cable failures cause Internet disruptions for multiple African countries
Cloudflare analysis of the March 2024 disruption pattern seen across African networks.
Resilient Internet connectivity in Europe mitigates impact from multiple cable cuts
Cloudflare case study showing how dense regional redundancy softened the impact of Baltic cable cuts.
A framework for measuring Internet resilience
Cloudflare research piece on route diversity, network topology, and the measurable ingredients of resilience.
Q2 2024 Internet disruption summary
Cloudflare’s broader review of internet disruptions in mid-2024, including cable-cut events.
The Impact of Submarine Cables on Internet Access
World Bank publication page for research on how submarine cable capacity affects affordability and access.
The Impact of Submarine Cables on Internet Access
World Bank research paper with the underlying findings on price effects and competition.
Resilient telecommunications infrastructure
World Bank paper on telecom resilience, natural hazards, and the value of alternative cable routes.
The January 15, 2022 Hunga Tonga-Hunga Ha’apai eruption and tsunami
World Bank rapid damage estimate documenting the cable break that isolated Tonga from the rest of the world.
Undersea cable fault could cut off Tonga from rest of the world for weeks
Reuters report on the operational difficulty of restoring connectivity after Tonga’s cable rupture.
Tonga reconnects to world as submarine cable restored after tsunami
Reuters report on the repair timeline and restoration of Tonga’s international cable link.
The Undersea Infrastructure Bringing More People Online in Emerging Markets
IFC briefing on how undersea cables lower prices, increase capacity, and improve digital access in emerging markets.
