Forecasting tornadoes in space
Solar storms striking Earth cause blackouts and costly damage. Can we protect ourselves?
• Extreme weather in space can affect us all on Earth – but hardly anyone talks about it. “Solar storms can harm satellites, interfere with GPS and air traffic, and damage power grids, or knock them out completely,” says Volker Bothmer, an astrophysicist at the University of Göttingen, in central Germany, who has been studying space weather for some 30 years.
Geomagnetic or solar storms involve vast numbers of charged particles being ejected during huge eruptions on the surface of the sun and racing towards Earth. In fact, they happen quite regularly. As recently as 2015 a geomagnetic storm halted all air traffic in Sweden for five hours when radar and GPS systems failed. In October 2003, Malmö, in Sweden, and South Africa suffered power outages. Germany has been affected, too, with flights being restricted because of disruptions to radio communications.
The biggest incident in the recent past was in 1989, when a transformer was destroyed in Quebec, Canada, resulting in a large-scale blackout. More than six million people sat in the dark for nine hours. “That solar storm was actually only of medium intensity,” says Bothmer. More severe ones, such as those belonging to the so-called Carrington class, can do substantially more harm. The Carrington category is named after the most powerful geomagnetic storm known to have hit Earth. In 1859, the British astronomer Richard Carrington observed two flashes of light on the surface of the sun, using his telescope; a few hours afterwards, polar lights were seen right down to the tropics and telegraph wires gave off sparks.
At the time, some 160 years ago, the world was not yet that vulnerable. But our planet has since become so electrically interconnected and wired up that some people warn the potential effects of a Carrington Event could be comparable to those of a nuclear war.
Bothmer, 57, would not go that far. He is a consultant for the European Space Agency (ESA) and Nasa, and is developing scientific tools to help predict the weather in space. Nevertheless, he believes the threat posed by adverse space weather is underestimated. “Here in Central Europe and Germany, we give too little thought to the potentially catastrophic consequences of solar storms,” he says. “We urgently need more accurate studies.” Countries such as Sweden, the UK and particularly the US are already carrying them out. Last year, a team of researchers headed by Edward Oughton at the University of Cambridge presented a detailed report describing what would happen if such a solar superstorm were to strike Earth.
The scientists concluded that large stretches of the North American continent would be propelled back to 18th-century conditions. Up to 215 million Americans might suddenly find themselves without electricity, depending which regions were affected, and power would not be restored quickly, because a severe solar storm could destroy the transformers. These serve as junctions within the power grid. They are custom made, the size of a family house and can weigh several hundred tonnes. Each one costs €5m-€7m (£4.45m-£6.25m) and takes at least five months to build. If hundreds of transformers were hit at the same time, it could take years, perhaps even a decade, for the country to recover, according to a study carried out by the US National Academy of Sciences in 2008.
Some regions would be left fighting for their lives, because refrigerators, ATMs and kidney machines would all cease to work, as would smartphones, petrol stations and supermarket tills.
Such a disaster could well cost $41.5bn (£31.6bn) a day, according to Oughton’s analysis. “Estimates of the potential socio-economic impact of a solar storm vary from country to country,” says Juha-Pekka Luntama, head of space weather at ESA. This is because assumptions vary with respect to the extent of the damage and the area affected. “We estimate that the effects of such an extreme space weather event in Europe would come to about €15bn,” says Luntama. “We deliberately took a relatively conservative view – if you wanted to, you could certainly paint an even bleaker picture.” However, Volker Bothmer warns that Europe and Germany are by no means safer than the US just because the potential cost of the damage appears to be lower. “Destroying a single transformer can lead to a major blackout,” as the solar storm of 1989 in Canada demonstrated.
Luntama agrees that a scenario of this type is possible in Europe: “Here, all the power grids are basically connected to each other. A regional outage could therefore theoretically spread here, too.” Both men urge every country to carry out stress tests on their power grids. Sweden and the UK have already done this. “Like the US, those two countries are now expanding the number of backup transformers kept in reserve,” says Bothmer. Emergency procedures have been clearly established too.
In Germany, the grid operators carried out their own tests and concluded that there appeared to be no problems, not even in the event of a severe solar storm. Bothmer wishes he could believe this, but says the power grid was studied during a period when the solar wind was not very strong. “That isn’t particularly useful,” he says.
“Even if bigger problems were to arise in Germany during a solar storm, we would still be able to respond to them,” says Mathias Fischer, a spokesman for the Dutch and German power grid operator TenneT. “Our crisis management is constantly analysing possible blackout scenarios.” Most of the measures that would have to be taken in the event of power outages due to a solar storm are much the same as those taken in the wake of cyberattacks, terrorist attacks or accidents. The power grid operators Amprion and TransnetBW take a similar view: they do not view solar storms as a separate issue. From the companies’ point of view, this makes perfectly good sense. They estimate the probability of a severe solar storm causing network outages to be vanishingly small. However, some experts doubt whether that assessment of the risk is correct.
Pete Riley, a space physicist at Predictive Science, a private company based in San Diego, California, concluded in an analysis published in 2012 that the probability of a Carrington class storm occurring by the year 2024 is around 12%. “That value is disputed among the experts; no one knows the exact figure,” says Luntama. “But I fear he could be right.”
So what is to be done? “Theoretically, many problems that arise from solar storms can be solved using engineering and technology,” says Luntama. Transformers can be made more resilient by installing duplicate components. Computer technology in satellites and aircraft could be screened better against damaging solar storm radiation using thicker casings made of aluminium, lead or tungsten. “But that is extremely expensive, and on top of this it makes the systems heavier and more unwieldy,” says Luntama. “So the question is, do we want to do this for something that is very unlikely to happen?” This question can only be answered using an individual cost-benefit analysis. “This will probably not be a blanket solution,” he says.
Luntama believes a far better compromise between financial investment and protection against possible damage could be achieved through better prediction of solar storms, which is what the 54-year-old is working on with his team at the moment. Until now forecasting space weather has been more like forecasting tornados, explains Luntama. “We can only say that there is a high probability of tornados forming in a certain region, but we don’t know exactly when or where.” This is because we have an incomplete understanding of the processes that lead to plasma eruptions on the surface of the sun. The unknowns in the calculations are still so large that it is not certain whether a solar storm will hit Earth or narrowly miss it.
The other side of the sun
Such predictions are too vague to justify taking expensive preventative measures such as switching off satellites, re-configuring power grids or re-directing aircraft. The problem with strong, fast storms is that once Luntama and his colleagues at ESA’s Space Situational Programme are sure that a plasma cloud is heading for Earth, there are only about 15 hours before it strikes. That is enough time for satellite operators, but airlines, energy suppliers and power grid operators would like two or preferably three days’ warning for their contingency plans. This would allow them to redirect flights more effectively and halt maintenance work on the power grids to distribute the energy more evenly. “That reduces the overall load – more power lines are available for the same amount of electricity. This makes the grid more robust,” says Luntama.
To extend the forecast from 15 hours to four or five days, scientists at ESA are pursuing an idea that Bothmer in Göttingen described in his doctoral thesis in 1993. Until now, astronomers have been observing the sun using satellites orbiting the Earth. So they are watching the side of the sun facing Earth. Bothmer suggests observing the side that will be facing Earth five days later. “That is crucial for a better forecast,” says Marc Scheper from the aerospace company OHB, based in Bremen, which is currently looking into the feasibility of a mission to the so-called Lagrange point L5 – where a space probe could be stably parked – for ESA. “That way we would see what was brewing, and that would give us a longer advance warning,” says Scheper. Another advantage is that looking from the side allows more accurate measurement of how fast the plasma fronts of a solar storm are travelling. “That will improve the temporal forecast,” he says.
The OHB engineer’s preliminary conclusion is: “The Lagrange mission should not be a problem to carry out, we are only working on the finer details.” If the European Council of Ministers, which is due to meet at the end of next year, approves the improved space weather forecasting programme, construction could begin by the middle of 2020. The satellite could then be launched from the European spaceport in French Guiana in 2023. Depending on the flight path, the first data could be available one or two years after the launch, according to Scheper. Bothmer and Luntama hope this will be the case. Luntama is confident that, “the issue of solar storms is gradually being taken more seriously”. This is also reflected in the budget for space weather forecasting, which has been increasing for some years now. At the moment, it amounts to some €40m. If the Lagrange mission gets the go-ahead, that figure will increase to about €500m.
For now, everything looks set for lift off. However, this is ultimately a political decision, according to Luntama. “Of course there is also the option of simply doing nothing and hoping that we stay lucky and are spared a severe solar storm. That would save us a lot of money. But if we are unlucky, we will suffer for it.” His opinion is very clear: “We should definitely be better prepared for Day Zero. It would be foolish not to be.”