EntryUpdated 2026/08/09

If the Cable Breaks, Can Satellites Take Over?

Partly — but they are not the same thing. Submarine cables carry almost all international traffic and satellites a small fraction of it. The difference is not only capacity but who decides when the service stops.

Read this first: How Do You Change the Status Quo Without Firing?

Cables fail in the water; constellations fail at the console

Carrier	What it actually carries	How it fails, how long to fix	Who can switch it off
Submarine cableThe bulk of international traffic	Almost all cross-ocean traffic. Everyday internet feels wireless, but the wireless part is only the last few dozen metres.	Mostly severed by fishing gear and dragged anchors, sometimes by seabed quakes. Repair needs a specialised cable ship — locate, lift, splice, relay — weeks when it goes well, longer with weather or permits.	Hard for anyone to switch off alone: owned by consortia of multiple carriers, so no single party can cut it — though the state hosting the landing station can, and landing points are few.
Geostationary satelliteBroadcast and remote areas	Broadcast, maritime and aeronautical links, remote-area connectivity. Efficient one-to-many, limited for interaction by latency.	Long-lived and rarely faulty, but one satellite covers a vast area — if it fails or is jammed, the whole footprint goes at once, with no spare to slot in.	The operator, plus the state that licenses it and its spectrum. With so few satellites, each decision affects a very large area.
Low-earth-orbit constellationA real backup, not an equivalent substitute	Traffic that is small in volume and high in importance: command links, financial settlement, emergency coordination, news. Not suited to an economy’s everyday internet.	A single satellite failing matters little, because there are many and they cover for one another. It also depends on no local physical infrastructure — the property that matters most in a crisis.	Typically one operator, able to enable or restrict service by geography from its own console. Nothing severs it — and service in a region can still end through a commercial decision, an export control, or the operator’s home-state policy.

“We have satellite backup” conflates two things: capacity and control. The first column is capacity — constellations suit traffic that is small in volume and high in importance, not an economy’s everyday internet. Treating them as an equivalent substitute for cable is the standard over-optimism here.

The last column is the decisive one, and it deliberately asks who can switch it off rather than who owns it. A cable’s fragility is physical: an anchor severs it and repair is slow. A constellation’s is governance: nothing severs it, but service can be shut off by geography from a console. The two do not substitute for one another — another cable does not solve a single-operator problem, and another satellite contract does not solve a seabed one. So “we have backup” needs the follow-up question: backup against which one?

Source: Impactful Creative, compiled from the layers described on this page and our coverage of the space economy and maritime grey-zone activity

How does international internet traffic actually travel?

Overwhelmingly through submarine cables. This surprises people, because everyday internet feels wireless — but the wireless part covers only the last few dozen metres. Crossing an ocean is almost entirely physical fibre.

There are fewer cables than people assume and far fewer landing points. Bringing a cable ashore needs suitable seabed, a protected stretch of coast, and a local landing permit, so several cables serving one region often converge on a handful of landing stations — a textbook single point of failure: many routes on paper, a few doorways in reality.

Satellites operate at an entirely different scale. Even the large low-earth-orbit constellations of recent years fall well short of a single modern submarine cable in total capacity — and that capacity is shared among every user across the coverage area.

How do cables break, and how long is the repair?

Most breaks are accidental: fishing gear and dragged anchors account for the great majority, followed by seabed earthquakes and landslides. Hundreds of faults occur worldwide each year and most go unnoticed, because traffic reroutes automatically.

Repair requires a specialised cable ship. There are not many, they are unevenly distributed, and the work means locating the fault, lifting the cable off the seabed, splicing it, and laying it back — weeks when it goes well, longer with weather, sea state, or permits. That is the sharpest contrast with terrestrial networks: on land you can dig the same day; at sea you cannot.

It is also why deliberate and accidental are hard to separate in the moment. The same marks can come from a trawler or from an anchor dragged on purpose, and attribution takes time and evidence — which is exactly why grey-zone operations favour this method: real damage, blurred responsibility, and a victim who must prove the case before responding.

What did low-earth-orbit constellations change?

Two things above all: latency and speed of deployment. Low orbits are far closer, cutting round-trip latency from hundreds of milliseconds in geostationary orbit to tens, which makes real-time calls and remote operation usable; and a terminal needs only an antenna, with no physical line at all.

That made satellites a genuine backup for the first time rather than merely an option for remote areas. In a crisis this is the valuable property: it depends on no local physical infrastructure, so it does not fail alongside that infrastructure.

The capacity gap did not change, though — it only became less obvious. Constellations suit traffic that is small in volume and high in importance: command communications, financial settlement, emergency coordination, news transmission. They do not suit carrying an economy’s everyday internet. Treating them as an equivalent substitute for cable is the standard over-optimism here.

They are also a textbook dual-use system: one constellation serving commercial users and military units at once, which complicates its position in a conflict — interrupting a commercial service and disabling a military capability become technically the same act.

Who decides when the service stops?

This is the deepest difference beyond capacity. Submarine cables are owned by consortia of multiple carriers, which makes unilateral shutdown hard. A constellation typically sits with a single operator, which can enable, restrict, or geographically limit service from its own console.

So resilience takes a different shape in each. A cable’s fragility is physical — an anchor severs it and repair is slow. A constellation’s fragility is governance — nothing severs it, but service in a region can end through a commercial decision, an export control, or the policy of the operator’s home state.

Which means “we have a backup” needs qualifying. Physical redundancy and governance redundancy do not substitute for one another: another cable does not solve a single-operator problem, and another satellite contract does not solve a seabed one. Real resilience here means having both and confirming in advance that they do not share a weakness.

The logic is exactly that of a chokepoint, except the choke is not a strait but a landing station and an administrative console: whoever holds the point holds asymmetric bargaining power.

Only the middle two can change state within minutes

Spectrum and orbital slotsReplaceable, over many years
Coordinated internationally and licensed by national regulators. It determines who may operate at all rather than who connects on a given day — a structural, long-run influence that does not shift during a crisis.
Launch capabilityReplaceable, and options are growing
Satellites have to get up there. Launch providers are few but multiplying — and what they affect is the pace of replenishment and replacement, not whether an existing constellation works today.
The operator’s consoleNo substitute — service can be toggled by geography
Activation, restriction, and geographic scope are all decided here — with no physical action and no advance notice required. This is the only point on the chain whose state can change within minutes.
The operator’s home-state policyNo substitute — controls apply to services too
Export controls and sanctions govern who may be served, not only what may be shipped. So even a willing operator can be required to stop — and that decision sits outside your contract with them.
Ground stations and terminalsReplaceable, and can be stocked in advance
Closest to the user and the easiest to handle: buy more, disperse them, test them beforehand. One of the few places where budget genuinely buys resilience — provided the four layers above are intact.

Replaceable, and can be stocked in advanceReplaceable, but measured in yearsNo substitute — takes effect with no physical action

The full answer to “who can switch it off” is not just the operator. Of the five layers, the outer three are all replaceable: spectrum and launch shape long-run structure, and terminals can be bought in quantity, dispersed, and tested in advance — one of the few places where budget genuinely buys resilience.

The problem sits in the middle two, which share two properties: no substitute, and no physical action needed to take effect. The operator’s console can enable or restrict service by geography; the home state’s export controls govern who may be served at all, so even a willing operator can be required to stop — and that decision lies outside your contract with them. Discovering this during a crisis leaves no time to arrange an alternative.

Source: Impactful Creative, compiled from the control points described on this page and our coverage of the space economy and the low-earth-orbit value chain

What does this mean for Taiwan?

Taiwan is an island whose international connectivity rests almost entirely on submarine cables, landing at a small number of points — which makes the single point of failure above concrete rather than abstract. It is starker for the outlying islands, where one break can drop connectivity to a minimal level, as has already happened.

Preparation works on two layers. Physically, more landing points and more route diversity, including deliberately indirect paths in different directions. In governance terms, ensuring backup does not concentrate in one operator, and negotiating crisis-time service conditions in advance — negotiate during the crisis and negotiating speed decides the outcome.

For an ordinary organisation, two questions are enough to take away: can my critical operations continue with connectivity fully down, and is my “backup” merely a second account on the same path? The second is the same question the supply chain piece asks — nominal diversification and physical diversification are frequently not the same thing.

Terms used here

Single Point of Failure — A node whose failure halts the entire chain. It is rarely the most expensive or most visible link — more often a specialty chemical, a sole packaging facility, or a particular test instrument accounting for a trivial share of cost. Mapping these is hard because they typically sit at tier three or four, while most companies have visibility only to tier one.
Chokepoint — A narrow passage through which global trade or energy transport must pass, with no equivalent alternative — the Strait of Hormuz, Malacca, Suez, Panama. The risk does not require closure: rising insurance rates, added voyage days from rerouting, and delay costs typically reshape trade economics well before any actual interruption occurs.
Dual-use — Technology, products, or know-how with both civilian and military applications. Dual-use character complicates export control determinations: whether the same chip, algorithm, or drawing is controlled depends on end use and end user — neither of which is typically visible to the exporter. Compliance burden accordingly shifts from product classification to customer due diligence.
Supply Chain Resilience — The capacity of a supply chain to keep operating through disruption and recover afterward, measured by the number of qualified alternate sources, inventory buffer depth, and switching time. Resilience trades directly against efficiency: every redundant supplier is another qualification and management cost. The real question is never whether to have resilience but which specific links justify paying the premium.