Loading

Black Swans

What if the network went down. Everywhere. All around the world. All at once. Think that’s impossible? It already happened – more than 160 years ago.

At the start of September in 1859, a huge flare detached from the Sun and began its journey across space, running headlong into the Earth’s strong magnetic field, trapping its highly-charged plasma particles in the magnetosphere. In the Northern Hemisphere, auroras could be seen as far south as Cuba (23º N). Those ghostly lights signaled the presence of a mass of invisible particles, each bearing a small charge. Like a lightning bolt from space, each particle sought to find a path to earth.

By the late 1850s many nations in Europe – along with much of the eastern United States – had already been strung with thousands of kilometers of telegraph wires. These turned out to be near-perfect receivers for those charged particles. Under attack from the flare, telegraph wires glowed blue with ‘St. Elmo’s Fire’, while telegraph operators received dangerous shocks from their equipment – which in more than a few instances burst into flames. In a widely-reported but nearly unbelievable bit of physics, telegraph equipment continued to function even after those operators had disconnected the batteries powering their telegraph stations, as the charge pouring in from space provided enough electric current to keep the system fully powered – but dangerous and unusable. While the solar storm raged throughout Earth’s magnetosphere, telegraphy ceased. It wasn’t until some time later that telegraphers realised these outages had been widespread – a global event.

British scientist Richard Christopher Carrington and another researcher, Richard Hodgson, independently observed that massive flare emanating from the Sun – the first solar flare in the scientific record. Today we call this the ‘Carrington Event’, and mark it as the starting point of the modern science of ‘space weather’.

In the decades that followed, we’ve learned a lot about the Sun’s 11-year cycle, charting how solar activity tends to peak and wane. That’s vital information for astronauts on the International Space Station, who occasionally need to shelter in the well-protected center of the ISS during a severe solar storm. We’ve also learned that solar flares can cause the upper atmosphere to expand – heated up by all those high energy particles colliding with it. An expanded atmosphere significantly increases atmospheric ‘drag’ on satellites in low orbits (which is most of them), potentially sending them tumbling unexpectedly back to Earth. Such a fate befell a squadron of 40 of Elon Musk’s Starlink communication satellites only last year.

More due to luck than anything else, Earth has not received another whopping head-on slam from a solar flare since the Carrington Event. Space weather researchers know that we can expect that the Earth will eventually be at the wrong place at the wrong point in the solar cycle, and we will get hit – hard – once again. It could even happen within the next few years, as we head toward the 2024-2025 peak of what already looks to be a very tempestuous solar cycle.

What happens then? We’ve come a long way from a few tens of thousands of kilometers of telegraph wires. Today, we have massive electricity distribution grids (acting as antennas and can be decimated by solar flares), tens of thousands of objects in orbit, and billions of devices that have inbuilt antennas – every smartphone, and much else besides. All of them will be impacted. Some of them will simply burn out, overwhelmed by the sudden charge. Some will simply stop working until the storm passes over. Many more will take subtle damage, never again quite working right.

The costs of a second Carrington Event could easily run into the trillions of dollars, as humanity worked to repair its electricity distribution grids, its disabled or dysfunctional mobile and wired communications infrastructure, while simultaneously needing to repair or replace most of its orbital communications satellites. Even those sums could pale in comparison to the global economic impacts of disrupted or destroyed communications networks. The past thirty years have seen humanity become deeply dependent on our networks. The pandemic showed us that even if we all stay in our homes, our communications networks will ensure that we can keep the civilisation humming along. Lose those networks, and we’ll lose the plot, economically.

But we needn’t look to space to spy a potential extinction-level threat for our networks. Cyberhackers – most often associated with nations such as Russia and North Korea – have become expert in developing the technology to penetrate deep into the core of our networks, disrupting them. Such disruptions will become more common – and more severe. But a disruption won’t necessarily be intentional: In late May 2023, Internet.NZ issued an apology after a “security mishap” blocked many .nz domains – taking thousands of businesses offline for many hours. Complex and distributed, with multiple points of failure and presenting many ‘attack surfaces’, our networks are both fundamentally vital to our economies – and surprisingly delicate.

So what can be done to prepare for the inevitable? Eventually, we will get that massive flare or massive cyberattack or massive own-goal that will shut our networks down for hours, days – possibly even weeks. Designed in the 1960s to be resilient in the face of nuclear war – keeping the military well informed of the status of its own and enemy forces as the world vaporised – the Internet of the 2020s looks fragile and increasingly creaky. Network engineers need to be given both permission and a mission: find ways to shore up their network’s operations, making them survivable in the face of almost any imaginable attack.

A decade ago, engineers at Netflix – which lives or dies on whether it can get its streaming entertainment out to its global base of customers – developed a simulation tool known as ‘Chaos Monkey’. Fire it up, and it wreaks (simulated) havoc on the network, pulling plugs, breaking equipment, reprogramming all of the accessible bits to do wildly unexpected things. Netflix used Chaos Monkey to ‘think the unthinkable’; as a result, its network has a reputation for resilience and fault-tolerance that its competitors envy. In a world where we can imagine the shape of this next Black Swan, we all need to have a deep think about how to bring the chaos in-house, in order to tame it. Before it comes a-calling.