Geomagnetic storms around Earth.jpg

All eyes on the Sun

Image credit: Science Photo Library

This year marks a solar centenary. In 1921, three waves of a major geomagnetic event wreaked havoc in different parts of the world. Have advances in science meant that lessons can be learned to avoid catastrophe if history repeats itself?

Over the course of three days in May 1921, a series of coronal mass ejections (CMEs) hit the Earth as three major geomagnetic storms.

During a geomagnetic storm, the magnetic field of the Earth fluctuates wildly over the course of a few minutes or hours. These fluctuations are larger closer to the Earth’s poles, but during very large storms, the magnetic field variation is wide and extends further.

During May 1921, towns and cities all over the world reported strange events. There were dazzling twilight displays of aurora borealis (created by electrically charged particles from the Sun) but at latitudes far lower than usual. Intense aurora sightings were reported in England and France, and as far south as San Antonio, Texas, and the island of Tongatabu in the South Pacific. On one morning observers reported at least five peaks of major intensity in a five-hour period across Europe and North America.

Electrical surges also caused damage to trans-Atlantic cables, interrupting shipping signals along the west coast of North America and the Philippines. There was disruption to telegraph traffic in Europe and North America, and to telephone exchanges. Sparks at one exchange caused a fire in Karlstad in Sweden and long-distance telephone lines in an exchange at New Brunswick, Canada, were burned out.

Conversely, there were reports of enhanced long-distance radio signals over the Atlantic and the Pacific, as enhanced ionisation resulted in stronger signals. New York stations received stronger signals from Berlin and Bordeaux, while links between stations in Samoa and Awanui, north of New Zealand, were recorded as “unusually good”.  

The first CME on 13 May around 1pm GMT was followed by a second, smaller one in the evening (around 7.30pm). In quiet periods, the magnetic field at the Earth’s surface is between 10 and 50nT. A paper by Mike Hapgood of the Rutherford Appleton Laboratory notes that the first CME pair, peaking at around 100nT, would have cleared the density in the region between the Sun and Earth, enabling the second and third CMEs to travel more quickly. The moderate magnetic activity may have pre-conditioned the Earth’s magnetosphere so that it responded strongly to following CMEs over the next few days.

A second CME on 14 May produced a sharp rise in intensity to +230nT over India and then waves of depression over Hawaii for two hours, with lows of -150nT. In Europe, the opposite was happening, with a field of waves that reached +350nT at their peak. 

The periods of magnetic waves were followed by seven hours of magnetic field activity so  intense that it could not be measured by contemporary instruments.

The intensity can only be estimated, based on the effects. A geoelectric field able to melt fuses in copper wires and cause a fire (as happened in the Swedish telephone exchange) would have needed to be 10V/km, which over a typical line length of 100-200km, resulted in line voltages of 1,000V.

On 16 May, a third CME occurred, causing intense activity, similar to the first event on 13 May.

One phenomenon that was particularly noticeable in 1921 was the effect of geomagnetically induced currents (GICs) on radio waves. The energy added to the ionosphere by the CME changes the way radio waves propagate; for example high-frequency (HF) radio is lost, because it uses the ionosphere to bounce the radio waves. In 1921, that meant interruption to information transmissions, communications, and radio broadcasts.

Today, CMEs can affect aircraft in the polar regions, which use HF radio for their communication systems. There may also be some risk from increased radiation doses to aircraft passengers and crew on long-haul flights, which take high-altitude flight paths. The charged particles may also cause the autopilot to disengage every few minutes.

CMEs can also affect satellites in orbit. If electrically charged particles enter the satellite’s computer systems they can cause ‘single event upsets’ disrupting operation.

Satellites are also used for global positioning systems (GPS). The GPS signal normally travels from a satellite straight to the Earth, but if the ionosphere has areas of different densities the radio waves are either diverted or slowed down. GPS relies on precise timing to establish a position in a navigation system, for example.

Today, aircraft and ships as well as smartphones and car navigation systems use GPS, which relies on timing of received signals for location and positioning. On some public transport systems, GPS tells the train if it is in the station and allows the doors to be opened. A mis-timed GPS signal could mean the train does not recognise where it is, and the doors stay locked.

Retail payment systems and cash machines also rely on a GPS signal to operate, so a GIC may temporarily affect shoppers. Financial trading, especially high-frequency trading, uses GPS to timestamp transactions, thus confirming the price. Any disruption or outage could result in suspension of financial trading around the world.

In the event of a large CME, emergency services may have to revert to back-up radio systems like VLF (very low frequency) or standard radio broadcast using radio transmitters. Local radio communications are not likely to be affected but navigation systems that rely on satellite signals will be. This could delay emergency responses but could also hamper road systems, which rely on satellite signals for traffic control, such as traffic lights or motorway systems.

One of the main effects of an electrical surge is the overload on the power grid, causing power loss. In 1989, a large solar magnetic impulse caused a voltage depression, triggering the Hydro-Quebec power station’s protection system to shut down, causing a blackout that lasted for nine hours.  

“What happens is the magnetic field of the Earth is varying very rapidly, and that variation of the magnetic field penetrates into the sub-surface of the Earth, which is slightly conductive,” explains Ciaran Beggan, senior geophysicist at the British Geological Survey. A changing magnetic field in a conductive material generates an electric field, like a generator with a rotating magnet inside a coil that generates electricity. The rapid changes over a large area can produce an electric field up to several volts per km. The field is harmless to humans, but in a high-voltage power network, the power lines have very low resistance, so that can transmit energy.

Transformers in the power grid step the voltages up or down and neutralise the additional electric field created by the magnetic storm. They balance the current load in the power lines but additional current from the ground, as a result of GICs, causes energy to leak out of the transformer. This heats up coolant oil in the transformer and when it reaches a critical temperature, the transformer automatically shuts down to protect itself.

According to Andrew Richards, modelling manager at National Grid ESO in the UK, the Armageddon scenario people imagine is unfounded. He believes the effects of a CME will be localised and “similar to what we’d see in a terrestrial storm, which can knock out power lines and leave small areas without power”. The structure of the grid in the UK makes it resilient. The size of the island means that it does not have very long power lines and there is redundancy built into the network.

This is not the case in North America where there is less redundancy and there are long stretches of power lines along the coastal areas. In the eastern half of the US, the power grid is highly interconnected; failure at one point could cause a domino effect, possibly resulting in power failures lasting days.  

The UK grid has back-up transformers, but if too many transformers are lost at one site, houses and businesses supplied by that electricity station could lose power. This is still an unlikely scenario, says Richards. What is more likely is that there will be a fluctuation in voltage, causing lights to dim.

Many associations, agencies and government departments are working together to identify any possible CME. There can be very little warning of impact, maybe 30 minutes from when the magnetic field position of the CME can be determined, to predict if it will cause an impact.

The focus of research is to identify the formation of space weather systems that can predict when solar flares are likely to happen. The Automated Solar Activity Prediction (ASAP) system, developed by Rami Qahwajii, professor of visual computing in the Faculty of Engineering and Informatics at the University of Bradford, detects, records, and predicts sunspot activity. It uses image processing and artificial intelligence to compare sunspot images provided by Nasa’s Solar Dynamic Observatory (SDO) with historical patterns to predict whether a sunspot is becoming more complex, likely to produce more solar flares or if it is no cause for concern.

The ASAP system is integrated into Nasa’s Community Coordinated Modelling Center portals, which shares data with all space weather watchers around the world.  

Predicting the timing and intensity of an event is a growing technology, as scientists learn from the Sun’s activity today and use historical data to try and gain as much notice as possible of the scale of an occurrence.

Solar flares happen all the time but only a few become significant CMEs that damage the Earth. Our reliance on electricity and satellite signals will make a significant geomagnetic event more disruptive than in past times, but parallel developments in technology have also implemented safeguards in the infrastructure, while co-operation between agencies means that advances in early warning systems are shared worldwide.


Lessons from history

1859: The Carrington event – Named after British astronomer, Richard Carrington, who made the first observation of a solar flare (also seen independently by Richard Hodgson) and linked the increased activity to the enormous geomagnetic storm of September 1859 – the largest on record. This storm is estimated to have ranged from -800 to -1,750nT in strength. It brought down telegraph systems in Europe and North America and caused northern and southern auroras to be seen in middle latitudes.

May 1921: A series of coronal mass ejections struck, reaching 350nT at the peak of geomagnetic activity.

March 1989: The Canadian province of Quebec was plunged into a power blackout for nine hours when the grid shut itself down within 90 seconds of being hit by a large solar storm and geomagnetically induced currents.

January 1994: The Anik-E satellites are thought to have been hit by a solar storm which damaged circuitry, causing them to lose Earth lock and to spin. The back-up systems failed to activate in time with the loss of data and Canadian TV transmission for several hours.

Satellite systems and communications were disrupted and aircraft rerouted flight paths to avoid high altitudes over the polar regions. Auroras were visible over the Mediterranean, Florida, and Texas.

February 2011: The Valentine’s Day flare was the largest solar flare in four years and prompted fears of increased activity that the sun was entering its solar maximum phase.

January 2012: Nasa’s Solar Dynamics Observatory observed a solar flare that measured M8.7 in intensity. Spacecraft recorded its speed at over 2,000km/s. The storm hit Earth but the magnetic activity was not strong enough to cause anything more than stunning auroras at high latitudes.

July 2012: A solar storm similar in magnitude to that in 1859 passed the Earth’s orbit but missed striking.



What makes a solar flare produce a geomagnetic storm on Earth?

The Earth’s magnetic field reverses once or twice every million years, but the Sun’s reverses every 11 years as part of its solar cycle. As the cycle develops, the ball of gas that makes up the Sun rotates faster than its poles, and the magnetic field underneath the surface can start to poke through.

“Sunspots and the magnetic field start to twist around each other and as the magnetic field twists into a loop, a little like an elastic band, it stores up energy. And when it ‘pings’, it releases a huge amount of energy in a very short space of time,” says geophysicist Ciaran Beggan. Energy can be released in a fraction of a second or in minutes, producing a very bright, white spot of light, or solar flare.

This solar flare is a huge amount of electromagnetic energy, light or gamma rays, which travels to Earth in just eight minutes. If it hits the day side of the Earth, it ionises the atmosphere, changing the way radio waves propagate to the ionosphere.

The release of energy pushes ionised gas away from the Sun in a coronal mass ejection (CME). The gas, or plasma, has the magnetic field it created embedded in it. “You can think of it as a cloud of rolling gas moving towards the Earth with its embedded magnetic field constantly rotating around,” says Beggan. This takes between 16 to 30 hours to reach the Earth, travelling at between 1,000 and 2,000km per second.

If the magnetic field in the CME is negative (pointing downwards) and the Earth’s magnetic field is positive, the CME injects energy into the Earth’s magnetic field creating a magnetic storm. “The energy from the CME is now interfacing directly with the Earth’s magnetic field and it creates a load of current systems in space that flow around the Earth. It also peels the Earth’s magnetic field off and around the back where it reconnects and then pushes more energy into the atmosphere,” continues Beggan. This creates electrical currents, or auroral electrojets, which generate a secondary magnetic field.


Sign up to the E&T News e-mail to get great stories like this delivered to your inbox every day.

Recent articles