It was the opening days of 2022, in the aftermath of a huge volcanic eruption, when Tonga went dark. The underwater eruption – 1,000 times more powerful than the bomb that was dropped on Hiroshima – sent tsunami waves across Tonga’s nearby archipelago and blanketed the island’s white coral sands in ash.
The strength of the Hunga-Tonga-Hunga-Ha’apai eruption severed internet connectivity with Tonga, causing a communication blackout at just the moment that a crisis was unfolding.
When the undersea cable that provides the country’s internet was restored weeks later, the scale of the disruption was clear. The lack of connectivity had hampered recovery efforts, while at the same time devastating businesses and local finances, many of which depend on remittances from abroad.
The disaster exposed the extreme vulnerabilities of the infrastructure that underpins the workings of the internet.
Contemporary life is really inseparable from an operational internet, says Nicole Starosielski, a professor at the University of California, Berkeley and the author of The Undersea Network.
In that way, it’s much like drinking water – a utility that underpins our very existence. And like water, very few people understand what it takes for it to travel from a distant reservoir to our kitchen taps.
Modern consumers have come to imagine the internet as something unseen in the atmosphere – an invisible “cloud” just above our heads, raining data down upon us. Because our devices aren’t tethered to any cables, many of us believe the whole thing is wireless, says Starosielski, but the reality is far more extraordinary.
Almost all internet traffic – including Zoom calls, movie streams, emails and social media feeds – reach us via high speed fibre optics laid on the ocean floor. These are the veins of the modern world, stretching almost 1.5 million km under the sea, connecting countries via physical cables which funnel the internet through them.
Speaking via WhatsApp, Starosielski explains that the data transmitting her voice will travel from her mobile phone to a nearby cell tower. “That’s basically the only wireless hop in the entire system,” she says.
From the cell tower, it will go through a set of terrestrial fibre optic cables, travelling at the speed of light underground. It will then go to a cable landing station – usually somewhere near the water – and from there to the bottom of the seafloor, before coming up at a cable landing station in Australia, from where the Guardian is speaking to Starosielski.
Spies, sabotage and sharks
That the data that powers financial, government, and some military communications is traversing cables not much thicker than a hose pipe and protected by little more than the seawater above them, has in recent years become a cause of concern for lawmakers across the world.
In 2017, Nato officials reported that Russian submarines had stepped up their surveillance of internet cables in the north Atlantic and in 2018, the Trump administration sanctioned a Russian company that was alleged to have provided “underwater capabilities” to Moscow, with the aim of monitoring the underwater network.
A Russian attack on undersea cables would cause “significant damage to our economy and to our everyday lives,” Jim Langevin, a member of the US house armed services committee said at the time.
The targeting of internet cables is a weapon that Russia has long held in its arsenal of hybrid warfare. When Russia annexed Crimea in 2014, Moscow severed the main cable connection to the peninsula, gaining control of its internet infrastructure, enabling the Kremlin to spread disinformation.
Global conflicts have also been shown to have unintended, disruptive effects on internet cable systems. In February, Iranian-backed Houthi militants attacked a cargo ship in the Red Sea. The eventual sinking of the Rubymar was likely responsible for severing three undersea cables in the region, which disrupted a significant portion of internet traffic between Asia and Europe.
The US and its allies have also expressed serious concern that adversaries could tap into the undersea cables to obtain “personal information, data, and communications.” A 2022 congressional report on the issue highlighted the increased potential of Russia or China to access the undersea cable systems.
It’s a method of espionage that the US is all too familiar with: in 2013 the Guardian revealed that the UK’s GCHQ had tapped into the network of internet cables to access vast quantities of communications between entirely innocent people, as well as targeted suspects. This information was then passed on to the NSA.
The documents, revealed by whistleblower Edward Snowden, also showed that an undersea cable that connects Australia and New Zealand to the US was tapped to allow the NSA to access Australian and New Zealand internet data.
Despite the raft of dangers and the increasingly vocal warning from western governments, calls for greater action in securing the cable network have largely gone unanswered and many see the threats as overblown.
“There are no publicly available and verified reports indicating deliberate attacks on the cable network by any actor, be it Russia, China, or a non-state group,” a 2022 EU report said.
“Arguably, this implies that the threat scenarios being discussed could be exaggerated.”
One expert speaking to the Guardian was blunter in their assessment, describing the threat of sabotage as “bullshit”.
The data bears this out, showing that sharks, anchors and fishing pose a greater threat to the global internet infrastructure than Russian spies. A US report on this issue showed that the major threats to the network are “accidental incidents involving humans”. On average, a cable is severed “every three days.”
“A submarine telecommunication cable was accidentally severed by a ship off the coast of Somalia in 2017, leading to a three-week internet outage costing the country $10 million a day”, the report states.
For many experts however, the greatest risk to the internet isn’t sabotage, espionage, or even rogue anchors – but the uneven spread of the cable infrastructure that threads across the globe, binding the world’s digital networks together.
“There aren’t cables everywhere,” says Starosielski. “There is a concentration in the north Atlantic Ocean connecting the United States and Europe but there are not that many in the South Atlantic.”
“So you see that some parts of the world have a high level of connectivity … and diversity in terms of having multiple routes in case there’s a break.”
As of 2023 there were more than 500 communications cables at the bottom of the ocean, but a quick glance at the map of the world’s undersea cable networks shows they are largely centred around economic and population centres.
The unequal spread of cables is clearest in the Pacific, where a territory like Guam, with a population of just 170,000 and which houses a US naval base, has more than 10 internet cables connecting to the island. New Zealand, with more than 5 million people has seven. Tonga has just one.
In the aftermath of the 2022 eruption in Tonga, governments across the world were spurred into action, commissioning reports into the vulnerabilities within the existing undersea cable network, while tech companies worked to bolster networks to ensure such an event never occurred again.
Last month, Tonga’s internet went down again.
Huge parts of the country were left in the dark after the undersea internet cable connecting the island network was damaged, causing chaos for local businesses.
For now, the economic fundamentals favour the building of more cables across the western world and into emerging markets, where the digital demand is booming. Despite the warnings of sabotage or accidental damage – experts say that without the market imperative to create more resilient networks, the real risk is that places like Tonga will continue to go dark, threatening the very promise of digital equity that the internet was founded on.