by Lucie Green
The problem is that the sites in the corona where the magnetic reconnection actually takes place are too small for us to see with current telescopes. The reconnection sites are probably less than 1 kilometre across, whereas our best coronal observations currently come from NASA’s Solar Dynamics Observatory, which can resolve down to structures 700 kilometres in size. So we are not yet able to see directly into the reconnection region, where the important physics is happening.
Still to be tested are the theories of exactly how the charged particles are accelerated from the corona to the chromosphere and photosphere. How does the energy actually transfer from the reconnecting magnetic field to the particles in the plasma? Is this occurring in the vicinity of the reconnection region because of electric fields that act as a particle accelerator? Or maybe it is occurring in the reconnection region itself? Another possibility is that turbulent motions are established – irregular flows that stir up the magnetic field and plasma, leading to changes in the plasma density and magnetic field. Observations on very small scales are needed to investigate. It is as important as ever to develop new and better ways of observing the Sun.
And there is still more to be learnt about the way the energy is transported to the lower solar atmosphere. Acceleration of electrons from the corona to the chromosphere is a mechanism that works, but it is not without some mysteries.
The number of electrons needed to generate the observed brightness of hard X-rays is much higher than the number of electrons that are in the vicinity of the reconnection region. To get the number needed, electrons need to be gathered from a vast volume of the corona. Either this is actually happening and our observations haven’t yet been able to reveal it, or we need to develop better theories. And one of my colleagues at the University of Glasgow thinks the latter is what we should do.
Lyndsay Fletcher is a solar physicist working in a group at the University of Glasgow, where there is a long history of studying solar flares and developing the theory to explain them. The idea that the electrons that stream down from the corona lose their energy by colliding with particles in the chromosphere was developed there. She and her co-worker Hugh Hudson developed a complementary idea whereby the energy is carried to the chromosphere not just by particles, but also by waves that ripple along the magnetic field lines. This is an interesting idea as during magnetic reconnection the shapes that the magnetic field takes on change, and waves can be created as they snap back into place. Like letting go of a plucked guitar string, waves propagate along the magnetic field lines. In a guitar, the wave energy can be transformed into motion of the air and generate sound. On the Sun, the wave energy can be transformed into an acceleration of the charged particles.
Importantly, if true, this energy transformation takes place in the chromosphere, where the coronal electrons have accumulated after their journey from the corona. This theory means that fewer electrons need to be accelerated down from the corona because they can be re-accelerated in the chromosphere as the magnetic waves ripple through. It’s a tantalizing idea. The hunt is now on to detect the waves and test the theory.
Back to what we do know: the last question is what causes the magnetic reconnection to start in the first place and trigger a solar flare? It could be that in the evolving magnetic field of the corona, magnetic fields get pushed up together and squeezed to breaking point. This produces a stand-alone flare. But often the situation is far more exciting.
I mentioned at the start of this chapter that the very early observations of solar flares showed that some of the plasma seen in the light of hydrogen alpha mysteriously started to move up as the flare went on beneath. This was initially undervalued by scientists and NASA, almost to their peril. It turns out that while solar flares may be the biggest explosions in the Solar System, above them are often the biggest eruptions …
12. Coronal Mass Ejections
‘Coronal mass ejections’ are the largest and most massive eruptions in the Solar System. The name describes an event when up to 10 billion tonnes of plasma blasts away from the Sun at millions of kilometres per hour. They can start relatively small – the size of a group of sunspots. But, as they leave, they expand to become many times larger than the Sun itself. They are beyond anything we can imagine. But, for something so awe-inspiring, they have been given a rather clunky, albeit descriptive, name. Yes, they are ejections of mass from the corona, but even the abbreviation ‘CME’ is nowhere near as catchy as their cousins: ‘solar flares’ and ‘solar wind’. Something like ‘solar eruptions’ would have been better; I’d even settle for ‘sun blasts’. But, no, CMEs it is, which makes them somewhat easy to disregard.
Putting nomenclature aside: a mystery even bigger than their name is how something that becomes so enormous went almost completely unnoticed until the 1970s – obscurity through implausibility, perhaps. With hindsight we can say that there had been hints though, and, for centuries, CMEs had been hiding in plain sight.
For example, if you look at sketches made during a total solar eclipse in 1860 you can see what is almost certainly a CME racing away from the Sun. Its circular shape makes it stand out next to the radial spokes that we call streamers. And once the spectroheliograph had been developed, hydrogen alpha light showed that the plasma in the corona did indeed form structures that sometimes erupted upwards. But it was assumed that this material never actually left the Sun.
12.1 Drawing of the total solar eclipse of 1860 by Wilhelm Tempel (Courtesy of the Royal Astronomical Society).
It was so inconceivable that plasma could be ejected in this way and travel out into the solar system that when NASA were planning the Apollo missions to the Moon they did not take anything of the sort into account. They only worried about high-energy particles that were somehow linked with solar flares. They had no idea that as a coronal mass ejection ploughs its way through the Solar System it can create showers of high-energy particles too.
It was actually in 1971 that we opened our eyes and saw CMEs for the first time. On 13 and 14 December 1971, images were being taken by NASA’s Orbiting Solar Observatory 7 which showed a section of the corona moving away from the Sun. Initially the technician thought there was an error with the camera, but it was soon realized that what they were seeing was a real event. In 1973 the head of solar physics, Richard Tousey, who had also recorded the Sun’s ultraviolet spectrum from a rocket for the first time, published what they had seen and the CME age began.
This is the era when I entered solar physics and it has been an immensely revealing and exciting one. The old approach of using solar eclipses to watch the corona has found new purpose and we have been confronted with the challenge of finding out where the energy comes from to power these massive eruptions. Along the way we discovered that they told us something very special about the Sun’s magnetic field. It is much more dynamic than we could ever have ever imagined and at times is like a pressure cooker, ready to explode.
FLYING MOUNTAINS
I said at the start of this chapter that CMEs involve a vast amount of material – as much as 10 billion tonnes of plasma. A big claim but one I can confidently make because we can calculate their mass thanks to the way in which we see them.
As we saw before, the corona is only seen in visible light during a total solar eclipse. With the photosphere blocked the corona is revealed because the electrons in the coronal plasma scatter the photospheric light. The corona is faintly visible because its electrons cause it to be effectively ‘lit up’ by the light from below. Some telescopes in space now have their own disc to replace the Moon to produce an ‘artificial eclipse’ but the process of viewing the
corona is exactly the same. So when we see images of a CME, we are looking at light scattering off the electrons in its plasma. We know there must be protons and other particles along for the ride, but they do not reveal their presence when this technique is used.
We can, however, calculate how many there should be. The brightness of the CME from the scattered visible light gives us a way to work out how many electrons there are. So we just need to work out, for every electron, how many other particles there are as well. From that we can work out the CME’s total mass.
Ignoring the trace elements that only appear in tiny quantities, roughly 10 per cent of the particles in the coronal plasma are helium and 90 per cent hydrogen. So for every helium particle, there are nine hydrogen particles. A hydrogen atom has one electron and helium has two, all of which have been stripped off in the plasma and are roaming free. So, our ratio of nine hydrogen for every helium will give us a total of eleven electrons. Helium nuclei are two protons with two neutrons, whereas hydrogen nuclei are a single proton. All said and done: for every eleven electrons we see, there must be eleven protons and two neutrons around too.
This is important to take into account because even though it is the electrons that are revealing themselves because they scatter photons, it is the protons and neutrons that give the CME its mass. The mass of an electron is nothing compared to these other particles. The mass of a proton is 1.6726 x 10(–24) g and the mass of a neutron is 1.6749 x 10(–24) g. The mass of the electron is about three orders of magnitude smaller than this. Sum all of this up and you find that there is an overwhelming amount of plasma in the ejection: somewhere between 10 million and 10 billion tonnes of material – roughly the same mass as Mount Everest.
You quickly see why CMEs were dismissed as unbelievable when you think about what it would take to blast something like Mount Everest off the Sun. This is only a few per cent of the mass lost from the Sun by the solar wind every day (the solar wind carries 1 million tonnes a second) and a negligible amount when compared to the mass of the Sun as a whole. But the huge gravitational pull of the Sun means that something really impressive is happening for a CME to be blasted away in one shot. I can’t even imagine it being blasted off the Earth. But I can imagine something smaller being sent off our planet.
What would it take to get a fairly small object, such as a tennis ball, to leave the Earth and never fall back again? We know from our own everyday experiences that the faster we throw the ball, the higher it goes. Yet, no matter how hard we try, as the ball rises, it always slows down before falling back to Earth. But, in theory, could someone with superhuman strength throw the ball fast enough for it to escape the gravitational pull of the Earth? Thinking in terms of energy, the ball would need to be given enough kinetic energy, due to its motion, to exceed the gravitational potential energy the ball has at the surface of the Earth. We need to work out the ball’s escape velocity.
A few equations can solve this for us. The kinetic energy* of the ball, indeed of any object, depends on its mass and the square of its speed. The gravitational potential† energy that the ball has is given by its own mass multiplied by the mass of the Earth and the gravitational constant – all divided by the distance the ball is from the centre of the Earth, the Earth’s radius.
Like a good algebra problem, the mass of the ball appears in both equations and so we can cancel it out. What is important for finding the escape velocity is the mass of the Earth and the distance that the ball is from the centre of the Earth. It does not matter how massive the ball is. The same speed applies to any object, whether a tennis ball or a cannon ball, which is to be blasted off the planet.
That speed is around 40,000 kilometres per hour or 25,000 miles per hour (depending on how you like your units). Ignoring any drag from the surrounding air, the ball would have to be travelling at 11.2 kilometres every second to escape, and we can follow the same logic to find the escape velocity for the Sun. We just need to scale up the size and the mass.
The Sun is enormous – beyond what we can imagine. We might have an idea of how large our local town or city is, and a vague sense of how big a country is, but the whole Earth is beyond what we can intuitively deal with. Well: over a million Earths would fit inside the Sun. Take a million of those things which are too big to understand, and that’s how big the Sun is. And its mass is vast too, giving it a gravitational pull at the photosphere that is twenty-seven times as large as the force of gravity that we experience on terra firma.
Obviously there are even bigger problems than gravity that would make you not want to be on the photosphere of the Sun. But even if the Sun had a cool, habitable surface, 27g would not be fun. That 100 grammes of corn chips we had before would be pulled down as if it were 2.7 kilograms on Earth. A normal 75-kilogram human would weigh the equivalent of over 2 tonnes. Besides that, sustained gravitational force of 10g is fatal for humans, so 27g would not give you long to enjoy your holiday.
The mass of the Sun is roughly 333,000 times that of the Earth and its radius is 109 times greater. So scaling up our previous calculation accordingly gives the escape velocity of the Sun to be 618 kilometres per second. But mysteriously that is not always the speed we see CMEs moving at.
One of the very first things I did in solar research was to look at CME speeds. Curiously, there is a large variation – from 20 kilometres to over 2500 kilometres per second. If you look into these numbers you find out that 50 per cent of the eruptions leave the Sun at less than the escape speed. This simple statistic told me something important about coronal mass ejections – they don’t need to reach escape velocity to escape. And this means they are not acting like a ballistic projectile. They aren’t like a bullet leaving a gun. There must be another way that these ejections are lifted off the Sun.
To come back to the Earth analogy, another way that the ball could escape the Earth’s gravitational pull is to be strapped onto a rocket so that the constant burning of the rocket fuel provides a continual force to overcome gravity and get into space. This is how the Saturn V rockets sent humans to the Moon and this is the approach that is relevant for coronal mass ejections: something must be continuously propelling them.
Well, ‘propelling’ does not do a CME justice. The Saturn V rockets were not able to send more than around 50 tonnes to the Moon (not including its own fuel). It takes a lot of energy to rocket something into space. CMEs not only have millions of times more mass, but they have the Sun’s immense gravity to contend with. Launching them would require a phenomenal amount of energy.
And it’s not hard to estimate just how much energy would be required. We know the mass and the velocity of CMEs from observations, so we can calculate their kinetic energy. Again, the kinetic energy is half the mass multiplied by the velocity squared. So, the kinetic energy of a CME carrying 100 million tonnes of material moving at 450 kilometres per second is of the order of 10 million billion billion joules. That is an insane amount of energy. That’s 10 septillion joules. Or: 10 yottajoules. That’s a yotta joules. Yet CMEs require at least that much energy to get them moving. This energy has to come from somewhere and this energetic mystery was the beginning of my career in solar physics.
COME FLY WITH CME
These ejections captivated me when I started to learn about the Sun and they pulled me in to want to understand more. By then, we had come a long way since the serendipitous CME discovery of the 1970s and there was now a spacecraft carrying a telescope dedicated to observing them. I was looking at images taken by coronagraphs carried on the SOHO spacecraft – artificial solar eclipse images made by placing a disc to block the light of the photosphere. Bu
t whereas a total eclipse gives a view of the corona for a few minutes, a coronagraph in space provides a view day after day and month after month: 14,000 CMEs were seen in the first three years of the SOHO mission alone.
Some of the ejections looked like amorphous blobs but some had a shape like a balloon with a bright centre. A particularly clear one with this shape was seen in the SOHO coronagraphs in 2000. Within the solar community it has become known as the ‘light bulb’ eruption because it had the shape of a glass bulb with a bright filament inside. (See plate 16.)
The SOHO coronagraphs are still being used today, despite the spacecraft having been launched in 1995. The SOHO coronagraphs take two to three images of the corona every hour, so they easily spot the eruptions and allow us to track them much further out from the Sun than had been possible with previous coronagraphs. These observations have been absolutely key to revealing how large these ejections become and that they are frequent events. There can be as many as six CMEs every day at solar maximum.
But despite the sudden wealth of observations of CMEs, we understood very little about them. I vividly remember being at a solar physics conference very early in my career and someone expressed frustration at how scientists had catalogued so many CMEs and could describe in detail what they looked like, but there was no detailed description of the physics actually causing them to happen in the first place. I decided that would be my first challenge as a solar physicist.
The main question about CMEs was the same question we had about solar flares: where is the energy to drive them coming from?