How New Zealand is preparing its electricity infrastructure for a rare but consequential threat

The first alert from NOAA's Space Weather Prediction Center arrived at 11.30am on 20 th May. It read, “WATCH: Geomagnetic Storm Category G4 or Greater Predicted.”

In the heart of Wellington, New Zealand, the team at Transpower, the country’s electricity grid operator, took notice. They assembled an Incident Management Team (IMT) led by Matt Copland, Head of Grid and System Operations. In a separate location, a panel of space weather scientists from Otago University, with access to data and observations of solar activity, did the same.

Less than hour after that first alert, the scientists, led by Professor Craig Rodger, informed the IMT that the coming storm was set to be a big one – a G5 category with intensities exceeding 1,200 nanoteslas (nT) per minute – and it would reach Earth within 12 hours. If left unchecked, an event this large could lead to days of electricity blackouts, and cause significant damage to the grid. With that information in hand, Transpower began alerting the wider electricity industry. They invited lines companies and generators to an emergency briefing, kick-starting a coordinated, national response to protect NZ’s electricity infrastructure, while keeping the lights on for as long as possible.

If you’re wondering how you missed the news of a devastating solar storm, it’s because it didn’t really happen.

I was lucky enough to sit in on Day 1 of Transpower's 2026 Industry Exercise ; a two-day live simulation designed specifically to stress-test New Zealand's electricity system against a major space weather event. The scenario was brutal by design – an extreme solar storm with the power to push the grid toward cascade failure. But it provided a valuable, yet safe, way to ensure that the infrastructure and the people who manage it can be ready for the real thing.

Our sun is incredibly dynamic – a ball of constantly-moving super-hot plasma and looping, twisted magnetic fields, which has 11-year cycles (and cycles over longer timescales). In addition to the light that all life on Earth relies on, our Sun also emits energy in more chaotic ways. We often see solar storms, bursts of activity akin to explosions, that can reach us here on Earth. There are two distinct types – solar flares and coronal mass ejections (CMEs). Both are hard to forecast and each has a different impact on our planet.

As US-based space weather scientist Dr Tamitha Skov explained to me previously , flares are “solar screams…they might be loud, but they don't physically hurt anything”. The strongest flares do nothing more than disrupt radio communications.

In contrast, CMEs are, Skov says, “a belching out of tons of material…it’s not just photons, it’s particles too”. These giant clouds of the sun’s plasma and its magnetic field can be expelled from the sun in any direction. Compared to light, they move relatively slowly. It takes about 8 minutes for sunlight to reach us, while CMEs take tens of hours (and the slowest can take several days).

Once a CME gets here, its magnetic fields can interact with the Earth’s. But as Craig Rodger explains, “The polarity of the CME makes a difference.” Speaking to me after the Industry Exercise, he continues, “If it is northward directed, it will rapidly compress the Earth's magnetosphere, but it’ll then flow around the planet and dissipate. If the CME is southward directed, it can couple to our magnetosphere. Instead of bouncing off, the CME pumps massive amounts of solar energy and particles into our magnetic system.”

This injection of energy and particles has one positive side effect - it creates the brightest, and most spectacular auroras in our skies. But its impact on the grid is very serious. A CME’s arrival induces voltage instabilities and powerful electrical currents, known as geomagnetically induced currents (GICs), through anything long and conductive…which is to say, the electricity network.

New Zealand's transmission network is unusually exposed. The country’s three main islands stretch between latitudes 34° and 47° south, placing them in the geographic range known to be prone to GICs during major solar events. NZ also runs roughly north-south, which aligns with the primary direction in which GICs flow. And significant parts of the grid sit on volcanic rock, which is highly conductive, making it easy for currents to enter the system.

In the case of a large CME – like the one in the simulation exercise – these GICs can overload the system, specifically impacting some of the many, many transformers that manage the grid’s voltage levels. Transformers are primarily designed to handle the alternating current (AC) that carries electricity throughout the grid, whereas GICs act more like direct current (DC). When DC flows through an AC transformer, it saturates the magnetic core, and produces intense, localized heating that damages the insulation, potentially causing catastrophic transformer failure. Or as Matt Copland describes it, “a permanent fault.” The transformer is, as he puts it, “kaput.”

For an organization or home plugged into the grid, a kaput transformer can mean power losses or voltage spikes that damage unprotected appliances. For grid managers, a major fault can overload other parts of the grid, causing cascading blackouts. And repairing or replacing a transformer is no easy task. The transformers used in electricity grids are custom-built, manufactured in small numbers by a handful of suppliers worldwide. “We carry some strategic spares,” says Copland, “but obviously not a huge number.” In a worst-case scenario – a catastrophic solar event that causes issues globally – there’d be a long queue of countries waiting to buy replacements. A reduced number of operating transformers could impact electricity supplies worldwide for months or even years.

There are ways to protect these workhorses of the grid. The first is to install a device called a GIC blocker onto key transformers. This blocker is effectively a capacitor, which acts as a sort of barrier to any unwanted DC currents, while allowing AC to pass through. Transpower will install a blocker at Benmore substation in the South Island, and more may be added in the coming years.

The other way to shelter an at-risk transformer from the ravages of a GIC is to temporarily disconnect it from the rest of the grid, removing the conductive path that the GIC needs to flow. Once the solar event has passed, the transformer can be reconnected, returning the grid to normal operations. While such disconnections require some short-term power outages – or in electricity industry parlance, “managed demand” – everyone I spoke to agrees that it’s preferable to losing a transformer permanently.

A yet-to-be-published economic analysis (commissioned by Transpower) which quantified the costs of mitigation vs. potential economic losses concluded that “in the absence of mitigation, a severe but realistic storm could result in up to NZ$8.36 billion in lost GDP.

It’s no wonder grid managers take this threat so seriously.

Switching And Recalculating

For me as an outsider, one of the most striking aspects of the simulation exercise was seeing the tension between uncertainty and action.

CMEs are very hard to predict in advance. But when one occurs on the sun, scientists can use solar-facing telescopes such as the Parker Solar Probe to estimate its magnitude and speed. With additional modelling, they can make predictions about its arrival time on Earth and its potential impact on the grid. But in terms of measurement, we have a big gap in our systems. Once the CME passes the dedicated solar spacecraft, it travels unobserved for many hours. “We only get more data on the CME – including its polarity – when it reaches the satellites at the L1 point , like SOHO [ Solar and Heliospheric Observatory ],”, says Rodger. “When you look at what Transpower and the electricity industry needs, it’s about two to three hours to get everything ready. With a fast-moving CME, they might only get 20 to 30 minutes from L1. That’s not much time.”

This meant that, as the scenario intensified, and the impact on the grid became clearer, exercise participants repeatedly had to make decisions based on best guesses or incomplete information. Based on Transpower’s switching plan, more than 50 organizations received indicative demand allocations: essentially a breakdown of how much power each of them would be allowed to draw once ‘rationing’ began. Each one had its own tab in a giant spreadsheet, and had to make plans about what to do with their allowance. In a ‘shadow’ control room at Transpower, in front of computer screens displaying the entire electricity grid, managers simulated the impact of these changes, reporting back to participants.

As the predicted size of the solar storm was increased (to 2,800 nT/min) several hours into the exercise, everyone had to pivot once again. Generators were asked to update their offers into the wholesale electricity market to reflect disconnections ahead of the storm. Lines companies were told to prepare for managed outages. The simulated media releases escalated from "we're monitoring the situation" to "power cuts will be widespread and could last up to three days."

The tempo was deliberate and intense, but – to my eyes at least – everyone remained calm and focused throughout.

A Scientific Support System

This isn’t the first time Transpower has considered the potential impact of geomagnetic storms on the electricity grid. The organization has been collaborating with Craig Rodger’s team at the Otago University for more than a decade. In 2015, Rodger established a research program called Solar Tsunamis , supported by government funding. It brought together experts from across NZ, the UK, Canada, the US and Switzerland, along with Transpower and Firstgas (who manage the national gas network).

One of the key outcomes of the collaboration was a validated model that can predict the geomagnetically induced currents that occur in the NZ electricity grid in response to solar storms of various magnitudes. Building this required 20 years of operational data from Transpower, solar observation data, and existing ‘scenarios’ used by the UK Met Office Space Weather team (because NZ and the UK are at very similar geomagnetic latitudes). This, in turn allowed Transpower to develop practical responses to space weather events – even up to extreme, 1-in-150-year events with storm magnitudes of 4,000nT/min. They then test their planned responses in simulations like the one I attended, or the one last November which considered impacts on systems beyond the grid, led by NZ’s National Emergency Management Agency.

This direct line between scientific research and operational decision-making is something that Rodger is clearly proud of, "We have a grid operator who take what we say seriously. When they’ve understood the potential impact of these events, they’ve used best practice to build plans to protect the key hardware. And through the exercises, they’re supporting the entire industry to test their tools in realistic conditions and get better prepared.” At the end of 2025, Rodger was awarded a follow-on grant, allowing the space weather collaboration to continue.

New Zealand's position in solar storm preparedness is, by most accounts, strong. Rodger puts it in characteristically direct terms. In some senses we could be doing better – the country lacks a dedicated government space weather forecasting body, which means relying on “phone-a-friend agreements” with the UK. But in others, "we are either ahead of everybody in the world or on the cutting edge." Dr Michelle Thaller, a former senior scientist at NASA, told 1News last year that NZ is the best prepared country in the world for space weather. Matt Copland's own assessment: "I think we are probably more prepared than most."

Speaking to me several days after the exercise, Copland says, “We definitely feel like we learned a lot. It’s very easy to have plans written down on paper – it almost gives you a false sense of certainty. But living through the interactions – between us and the science panel, and between us and the industry participants – it was very instructive.” In practical terms, a lot of the response, he says, is “balancing supply and demand, so managing that with the lines companies is critical. This was the first simulated test for a lot of the participants, so it was really about education and engagement. And it went well.”

While all of this preparation is happening on Earth, the sun is continuing to do its own thing. Solar Cycle 25 (which started in December 2019) has been more active than many expected. But according to heliophysicist Tamitha Skov, we're simply returning to historical norms after an unusually quiet previous cycle. The sun, she notes , still sits just below the median of all recorded cycles. There is, in other words, room for things to get significantly…. stormier.

Given that, I found it genuinely reassuring to see an exercise like this up-close. To sit in a room full of grid operators and emergency managers, as they liaised with scientists, lines companies, generators, retailers, and large electricity users all over the country; united in an effort to prepare our electricity grid for a solar storm that would otherwise leave us in the dark.