CONCEPT Cited by 1 source
Submarine cable incident¶
A submarine cable incident is damage to or failure of an undersea fibre-optic cable that carries a non-negligible share of a country's or region's Internet connectivity. Submarine cables are the physical substrate of the transoceanic Internet: when one fails, the traffic it carried must reroute onto redundant cables and terrestrial bypasses — if those exist with sufficient headroom. When they don't, national-scale connectivity loss follows within minutes.
Causes¶
Submarine cable failures have four recurring causes:
- Anchor and fishing-gear damage — ships dragging anchor or fishing trawlers snagging the cable, the single largest failure class globally.
- Landslides and seismic events — subsea earth movement shears the cable.
- Equipment failure at the cable landing station (CLS) — power, optical, or SDH / OTN / ROADM kit failure on the terrestrial side.
- Deliberate sabotage — rarer but occurring, particularly in geopolitically tense maritime corridors.
Observational signature¶
Submarine cable incidents produce a distinctive signature on Cloudflare Radar country graphs:
- Sudden drop in country traffic (often 50-90% below baseline) coincident with the cable failure timestamp.
- Not a BGP withdrawal — prefixes often remain announced; the routes simply have no working physical path.
- Provider-specific impact — traffic from ASes that relied on the failed cable drops hardest; ASes with redundant paths absorb more of their load.
- Recovery timescale matches repair complexity — from hours (landing-station kit reset) to weeks (ship-dispatched subsea splice repair).
- Provider disclosure matches the engineering reality — operators post on social media ("an international incident on the WACS cable") and announce backup activation within minutes-to-hours.
The Republic of Congo / WACS case (canonical instance)¶
Just after New Year 2026, the West Africa Cable System (WACS) — the ~14,500 km submarine cable along Africa's western coast connecting several sub-Saharan countries to Europe — sustained "an international incident." The downstream effect in the Republic of Congo:
- Cloudflare Radar observed a sudden drop in traffic from Congo starting at 00:00 local on January 2 (23:00 UTC on January 1).
- Traffic fell to 82% below expected levels.
- Congo Telecom (AS37451) acknowledged the incident on X and announced that "backup solutions had been activated" — the activation of terrestrial / satellite / alternate-cable paths not normally in the primary traffic path.
- Users experienced "slowdowns during peak hours" during the rerouted-traffic window — redundant paths had less headroom than WACS's primary.
- Full recovery at approximately 15:00 local on January 4 (14:00 UTC), ~63 hours after the initial drop.
This is the canonical shape: a multi-day national-scale disruption driven by a single submarine cable's failure, with the recovery pathway managed on-country by the dominant operator rather than by any global action.
Why submarine cables still matter¶
Despite the growth of satellite Internet (LEO constellations: Starlink, OneWeb, Kuiper) and terrestrial bypass:
- Cost per bit over submarine fibre remains cheaper by 2-3 orders of magnitude than satellite for bulk transit.
- Latency remains lower than geostationary and competitive with LEO for short-distance hops.
- Capacity scales far higher — single cables carry tens of Tbps versus tens of Mbps per LEO beam.
Consequently, most countries' bulk traffic rides cables, with satellite as a backup or niche path — and cable incidents continue to produce large country-scale disruptions.
Seen in¶
- sources/2026-04-28-cloudflare-q1-2026-internet-disruption-summary — canonical wiki instance; the January 2-4 WACS cable incident cut Republic of Congo Internet traffic 82% below expected levels for ~63 hours. Congo Telecom activated backup solutions; users experienced peak-hour slowdowns during the rerouted window.