Atlas Gridwatch INTELLIGENCE_LAYER // PUBLIC
← Return to Briefing
Published: 2026-01-04 | Classification: OPEN SOURCE

Global Undersea Cable Chokepoints and Landing Clusters: A Strategic Risk Assessment

Introduction: Digital Lifelines and Hidden Chokepoints

Modern civilization runs on a vast undersea cable network that carries over 95-99% of intercontinental data traffic 1 2. These fiber-optic "digital lifelines" underpin everything from high-frequency financial trades and SWIFT banking transactions to cloud computing and AI workloads. Nearly $10 trillion in daily financial transactions and even latency-sensitive AI interactions depend on subsea cables 3. Despite their criticality, this infrastructure remains largely invisible and surprisingly vulnerable. Cables—typically only a few centimeters thick—traverse oceans and converge in narrow passages and coastal landing sites, creating geographic chokepoints where disruptions can have outsized ripple effects.

Physical and geopolitical factors funnel many cables along the same routes - for example, through narrow straits and clustered landing stations - leading to single points of failure in the global internet. Brief cable outages have slowed stock exchanges and payment systems 1, and major cable cuts have abruptly knocked entire regions offline. This report provides a comprehensive analysis of undersea telecom cable density, chokepoints, and landing station clusters, identifying global patterns and high-risk zones. We examine redundancy (or lack thereof) in these networks, review past failures (from the 2008 Mediterranean cuts to the 2023 Red Sea attacks) and their impacts, and assess emerging risks posed by new infrastructure projects (e.g. Chinese vs. U.S.-backed cables and private hyperscaler systems). Finally, we consider how this submarine cable web underpins global cloud and AI infrastructure, and present risk matrices by region to evaluate resilience and vulnerabilities.

Global Cable Density and Major Routes

Undersea Cable Scale & Distribution: As of 2025, over 570 active submarine cable systems span ~1.4-1.5 million kilometers of ocean, with roughly 80 additional cables under construction 4. These cables land at more than 1,700 landing points worldwide (up from 1,444 in 2023 to 1,636 in 2024) 5. Importantly, cable landings are unevenly distributed: there are dense clusters in regions with major data hubs—notably Asia-Pacific, Europe, and North America - while transit regions like the Middle East and East Africa host critical junctions for intercontinental routes 6. Rather than a mesh covering all coasts evenly, cables concentrate along a few high-capacity "trunk" routes. The North Atlantic, for instance, carries massive traffic between North America and Europe; similarly, the Mediterranean-Red Sea corridor links Europe with Asia via the Suez, and the Strait of Malacca links the Indian and Pacific Oceans 7. This network design reflects efficiency and geography - but it creates systemic vulnerabilities, since vast volumes of data rely on a small number of physical pathways 7.

Physical Chokepoints: Many cables are funneled through tight maritime chokepoints and narrow straits due to geography. Notable global chokepoints include:

Other notable pinch-points include the Luzon Strait (between Taiwan and the Philippines, where multiple Pacific cables converge and where undersea earthquakes in 2006 famously cut several cables at once) and various passages in the Indonesian archipelago that intra-Asian cables thread through. Geographic chokepoints concentrate risk: as one U.S. Homeland Security analysis noted, while major incidents in such chokepoints are relatively rare, each underscores the vulnerability of regions where cables are similarly concentrated 12 13.

Landing Station Megaclusters: Beyond narrow straits, cable landing stations (CLS) in certain coastal cities have become megaclusters, hosting many cables in one locale. These clusters arise near major internet hubs where data centers and terrestrial networks converge. Key examples:

Clustering at landing stations is a double-edged sword: it creates efficient interconnection points, but also means damage or attacks at one site can knock out many cables at once. Landing stations supply power to cables and tie into terrestrial grids 21; yet they often were chosen for ease of permitting or proximity to infrastructure rather than physical security 22. As a result, cable landings often sit in unremarkable coastal facilities that could be vulnerable to sabotage (e.g. a well-placed explosive or power cut)—a risk recognized by European cybersecurity agencies 23. For example, a 2022 incident in Marseille saw multiple cable cuts (suspected sabotage) at once, illustrating the cluster risk 17.

High-Risk Regions: Geopolitical and Physical Vulnerabilities

Certain regions stand out as especially high-risk for cable disruption, whether due to geopolitical tensions, conflict, natural hazards, or simply the lack of alternate routes. Below we analyze several hotspots:

Summary: Regions combining dense cable convergence and high threat levels (like the Red Sea, Taiwan/SCS, and parts of Europe's periphery) present the greatest systemic risk. In many cases, redundancy on paper proves insufficient in practice—e.g. having 4 cables in one trench is no help if an anchor or landslide cuts all four together. The next section examines how redundancy (or lack thereof) plays into resilience.

Redundancy vs. Single Points of Failure

Global network resilience hinges on route diversity and spare capacity—effectively, having alternate pathways when a cable fails. In well-connected regions (e.g. North Atlantic), a cut to one cable usually results in traffic re-routing automatically through others, often with users none the wiser. Indeed, an estimated 150-200 cable faults occur annually 55, yet most go unnoticed by the public because operators engineer alternate routes 56. However, where redundancy is limited, a single incident can cause major outages.

Clustering and Co-Routing: A major concern is that many supposedly "separate" cables actually follow the same route or land at the same station, creating latent single points of failure. As noted, multiple cables often cluster through geographic chokepoints or single landing stations, meaning one accident can take out several at once. For example, the February 2024 Red Sea incident saw three independent cable systems cut by one ship's anchor 27. In May 2024, a separate cut of two cables (EASSy and SEACOM) in South Africa simultaneously dropped all East Africa-South Africa connectivity 57. In such cases, having multiple cables didn't help because they weren't diverse - all ran through the same bottleneck.

Lack of Alternatives: Certain countries or regions rely on just one or two main cables—true single points of failure. For instance, Mauritania had a single international cable (ACE) until recently; when it was cut in 2018, the entire country lost internet for two days. Many Small Island Developing States (Pacific islands, parts of the Caribbean) also depend on a single subsea link. Even at a continental scale, Africa's connectivity long hinged on only a few routes up North and around South - recent cuts have exposed weak points (e.g. West Africa's reliance on the WACS cable, which when cut in 2020 and 2021 severely reduced bandwidth to multiple countries). Vietnam in early 2023 was another stark example: all five cables cut = 75% capacity loss, no full backup for months 43.

Redundancy Measures: Telecom operators are increasingly investing in route diversity to avoid these scenarios. For example, after the Feb 2024 Red Sea cuts, African carriers like Kenya's Safaricom announced they were activating alternate routes and capacity on unaffected cables (like the TEAMS cable via a different route) to keep data flowing 58 59. Many countries now mandate or encourage operators to use at least two different submarine routes. However, true redundancy is costly and not always possible given geography and politics. Permitting constraints mean cables often end up laying in the same narrow corridors (e.g. Red Sea, Malacca), undermining physical diversity 60 61.

Role of Cable Maintenance Capacity: Even when alternate paths exist, limited repair capacity can turn a minor fault into a prolonged outage—effectively a single point of failure in time. There are only ~80 dedicated cable ships in the world 62, and repairs need skilled crews and favorable weather. If multiple cables break at once, or in wartime, repair delays can stack up. Notably, repair times have been rising - the average restoration time reached ~40 days in 2023 63, up from ~20 days in prior years. In remote regions like Africa or South America, a single available repair ship might have to triage multiple breaks. The Insikt (Recorded Future) analysis found that the worst recent outages (Red Sea, West Africa, South Africa 2024) all combined limited redundancy, co-located cables, and slow repair logistics 64 65. In other words, it's the intersection of these factors that causes systemic failure.

Consequences of Single Points: When redundancy fails, the impacts can be severe. Beyond slowing consumer internet, real economic damage occurs: e.g., the 2008 Mediterranean cable cuts (SEA-ME-WE-4 and others near Egypt) disrupted service to 75 million users from the Middle East to South Asia 66. In that event, 60 million in India lost or saw degraded connectivity, along with 12 million in Pakistan and 6 million in Egypt 67 49. International trading and outsourcing were hampered for days. Financial networks like SWIFT, which rely on global data links, experienced delays - at one point in 2008 some banks had to revert to backup satellite links (with high latency and low bandwidth) to process transactions. In 2016, a cable cut near Hainan isolated multiple Chinese cities from the global internet for half a day, illustrating even major economies can be caught off-guard by a single break if routes are insufficiently diverse.

In summary, true resilience requires both physical route diversity and fast restoration capability. The reality, however, is that many global regions still have obvious single points of failure. The table below summarizes a few examples:

Region/Route Chokepoint / Single Point Redundancy Level Impact if Cut
Europe-Asia (via Egypt) Suez & Red Sea chokepoint (all cables) 25 Low - no alternative route (all traffic funnels here) 24 Huge ~70% Europe-Asia traffic affected by one incident 32; multi-continent outages
Taiwan Limited cables (few routes via Luzon Strait) Low - 2 main southbound cables often cut 36 High - island could be isolated; 75%+ capacity loss if both cut
South China Sea Many cables but contested waters Moderate alternate routes far around (via Pacific) High - frequent minor cuts; conflict could sever all, causing major intra-Asia latency increases
West Africa Landing clusters (e.g. Abidjan) Low - few cables, often co-located 68 High - multi-country outages; e.g. 2024 slide cut all 4 cables off West Africa at once
US-EU (N. Atlantic) Multiple cables, clustered landings High capacity, but clustered landings (UK/US) Moderate - capacity allows reroute, but sabotage at landing could hit several cables
Pacific NW - Japan Multiple cables in single seismic zone Moderate routes close together across Pacific High - major quake/tsunami could cut most Trans-Pac links at once, isolating regions temporarily
Small nations (islands) Often a single international cable Very low - satellite backup only High - complete outage of internet/voice until repaired (weeks)

As the table suggests, a lack of route options (Egypt, Taiwan, islands) or co-location of "diverse" cables (West Africa, Red Sea) leaves several high-impact single failure points in the global network.

Historical Cable Failures and Outages: Lessons Learned

Examining past cable failure incidents helps illustrate the real-world impacts and drive home the importance of resiliency. Below are case studies of major outages and their consequences:

Impacts on Finance and AI/Cloud: These incidents have shown concrete impacts on critical services:

In summary, past failures teach that redundancy, diversity, and rapid repair are key. They also highlight emerging threats (intentional attacks) on top of the usual accidental breaks. The next section looks at future infrastructure projects in this context—will new cables reduce these risks, or could geopolitical rivalries create new choke-points?

Emerging Infrastructure Projects: Geopolitics of New Cables

The undersea cable landscape is entering a new phase of growth and competition. Dozens of new cables are planned through 2026, many backed by either Chinese or U.S.-aligned interests, and tech giants ("hyperscalers") have become dominant investors 75. These developments carry both opportunities for improved resilience and new strategic risks:

In summary, new cables are adding capacity and alternate routes - which is positive for resilience - but the political bifurcation of cable consortia (China vs West) could introduce strategic vulnerabilities. If the internet bifurcates, countries may have to choose which cable network to align with, and adversaries might target cables as part of larger conflicts. The ideal outcome is that more cables = more redundancy for all; the risk is that more cables just create two siloed networks, each with its own chokepoints.

Strategic Significance for AI, Cloud, and Global Stability

Submarine cables are often described as the "physical backbone" of the internet—an almost invisible infrastructure that makes our digital, cloud-driven world possible. The stability of this cable network has direct implications for global economic and security stability, especially as we enter an era of AI-driven services and ultrafast communications. Key strategic considerations include:

Finally, risk matrices can help prioritize where investment and protection are most needed. One way to score risk is by combining hazard probability, choke severity, and impact magnitude. By that measure:

These assessments underline that ensuring resilience of submarine cables is now a critical strategic priority for nations. The stakes - from everyday connectivity to the stability of financial and AI systems - demand a coordinated international approach to secure these undersea arteries. Robust investment in redundancy, cooperative security measures, and emergency preparedness will be needed to prevent the next cable crisis from spiraling into a broader economic or security catastrophe. As one expert noted, the world's submarine cables were "built for an era of cooperation, not conflict" 95—adapting them to today's realities is an urgent task for both industry and governments.

Sources

Generated by Atlas Gridwatch Research Engine