15 Million Degrees

by Lucie Green

  So that is it in a nutshell: the Sun transfers energy to the Earth in a range of different wavelengths, and a change in the magnetic activity on the Sun can have a knock-on effect which means that different regions on the Earth can be affected. But the impact of this is so localized that not only can it not cause an Ice Age, but also it cannot even compete with the human-made climate change that is dragging the temperature the other way.

  So, that makes the headline wrong. A drop in sunspot numbers is not going to plunge the Earth into a mini-Ice Age. But what about the claim that a ‘rare drop in sunspot activity’ is happening?

  SPOT ON

  During the years around 2009 the Sun really was unusually quiet in terms of the number of sunspots that appeared. And this lull took most by surprise. Not Mike Lockwood and his collaborator Claus Fröhlich though. They had noted that the long-term solar activity levels had been going down since 1985 and that the trend was likely to continue. So what was happening?

  As we saw before, the Sun’s magnetic activity waxes and wanes over roughly an eleven-year cycle, which is marked by an increase and decrease in numbers of sunspots. Once the knowledge of a cycle was established, the data were backdated so that cycle 1 started in 1755. Solar cycle 23 had begun in 1996 and reached its maximum in the year 2000. Using the average cycle length of 11.1 years predicts that the Sun would be back in a minimum phase again by around 2007 – ending cycle 23 and starting solar cycle 24. Using the experience from the past cycles, scientists tried to predict what cycle 24 would have in store. When will the cycle 24 start and how large it will be?

  To do this, the American National Oceanic and Atmospheric Administration, with support from NASA, created the ‘Solar Cycle 24 Prediction Panel’. This panel gathered expertise from around the world and made use both of observations of the Sun and of models of solar activity. Several teams had to be brought together because a solar cycle prediction can be made in several different ways.

  There are three ways to predict how a solar cycle is going to pan out, with increasing amounts of warning being traded off for accuracy in the predictions. If you wait for the cycle to actually start, the pace at which the sunspots appear can be an indicator of what will happen in the rest of the cycle, but by then the cycle is already under way. A step back involves looking at the conditions in the Sun before the cycle starts and seeing what information is there. Even further back, historical sunspot data, which span several cycles, can be used. But this method tries to tease out longer-term trends.

  The most common prediction method is perhaps the most obvious and falls into the first of the three approaches: considering each solar cycle as an individual unit of the Sun’s activity, with little interaction between cycles. Using this approach, previous sunspot cycles can be used to see whether there is a relationship between a cycle’s sunspot number and the time and size of its maximum.

  By studying lots of solar cycles it has been discovered that those cycles which show a rapid increase in the number of spots early on are likely to have large maxima, which arrive quickly. A sluggish start to the appearance of sunspots means a longer wait for maximum. This is known as the Waldmeier effect after the director of the Swiss Federal Observatory in Zurich who discovered it. The problem is that no prediction can be made until the spots of that cycle appear. After that we can of course fine-tune the prediction as more sunspots appear, but it would be helpful to have more warning about what a solar cycle is going to be like. For that, we have to look beyond just sunspots.

  Another approach to solar cycle prediction falls into the second category and uses the Babcock–Leighton scenario, which looks at the magnetic field lines running between the poles of the Sun (the Sun’s ‘poloidal’ magnetic field) because this is the seed magnetic field from which sunspots will later be produced. In theory, the poloidal magnetic field should be able to provide clues about the strength of the next cycle. Unfortunately, to cut a long story short, it is really difficult to actually track and measure the poloidal magnetic field. And whereas observations of sunspots go back 400 years, suitable observations of the magnetic field at the poles of the Sun really only go back to the 1970s.

  Then there is the flow of the plasma itself, which is what drags the magnetic field around and amplifies it in the first place. The ‘meridional flows’ are the movement of plasma between the equator of the Sun and the poles. We can see these flows on the surface of the Sun transporting magnetic flux away from the equator, and we know that deep within the Sun the returning flows take the field back down again. And despite their slow nature, these flows represent a circulation of the solar plasma that appears to be involved in driving the solar cycle. It makes sense that because buoyant plasma then brings this magnetic field back up from the tachocline (after it has been strengthened by flows dragging the field around the Sun) to the surface to form sunspots, changes in the meridional flows providing this magnetic flux could have knock-on effects for the number of sunspots eventually emerging. But now there are so many factors and variables involved that the predictions become much more difficult.

  But you don’t just have to focus on the Sun. After all, we know that the consequences of its evolving magnetic field reach out into the Solar System. This has meant that one approach to predicting the size of a solar cycle, which has had some success, uses the level of geomagnetic activity, the disturbance to the Earth’s magnetic field caused by the solar wind. This work has shown that there is a good correlation between the amplitude of a solar cycle and the level of geomagnetic activity at the previous solar minimum. That level of geomagnetic activity is set by the solar wind and the magnetic field embedded within it at the solar minimum and is thought to depend, to some degree, on the activity that took place in the preceding solar cycle.

  This has given rise to a different approach to solar cycle prediction, which is to consider solar activity over timescales much longer than a single solar cycle. The long-term data are then taken as being continuous and periodic variations in the data are looked for. To take the weather around you on Earth as an example: every day it starts cold in the morning, gradually warms up during the day and then cools down again at night. The twenty-four-hour cycle of the temperature around you is a bit like the eleven-year sunspot cycle. If you look at more data you will see that not only does the temperature vary across a day, but that those cycles themselves gradually get warmer and cooler during the year (the seasons) and even those seasons can vary in the very long term.

  In the sunspot numbers there is the eleven-year cycle, but superimposed on top of this cyclical variation are other periodic changes that occur over different timescales, such as the roughly eighty-year Gleissberg cycle (named after Wolfgang Gleissberg, who first suggested a periodic variation over this timescale – we’ll come back to this later). If the collection of periodic changes can be picked out from the number of sunspots, they can be used to extrapolate into the future to give a probability of the size of future cycles before they have even begun.

  Using this range of different techniques, the first long-range forecasts for cycle 24 were issued in 2006. A forecast at this time was difficult because the previous cycle hadn’t yet ended, no cycle 24 spots had emerged to give an indication of the cycle progression. Despite the challenges, it was predicted that solar minimum would occur in March 2008 and the maximum of cycle 24 would be reached in 2011. The predictions forecast a cycle that would be moderate to large in size.

  In 2009, a ripple of excitement started to flow through the solar physics community. The first spots of cycle 24 had been seen in January 2008. We knew they were from cycle 24 because they were of the reverse m
agnetic orientation (remember, it flips every cycle). Sunspots of cycle 23 that appeared in the northern hemisphere had a positive magnetic polarity in the leading spot, whereas those of cycle 24 are negative. The sunspots were given the name ‘10981’ (sunspot pairs are numbered in a sequential way), an inconsequential name but one that now represents a change of season on the Sun.

  Within three days the spots were gone though, as the magnetic field was quickly dispersed into the surrounding photosphere. Even though the sunspots were small and only lived for a few days, they were important and signalled that cycle 24 was here. For a while the spots of cycle 23 and 24 overlapped. It’s completely normal for the Sun to do this. Eventually the spots of cycle 24 began to dominate over the spots of cycle 23 in September 2008.

  Further cycle 24 spots, however, were very slow to appear and this is why the solar community grew excited. The Sun seemed to be quieter than it had been for the previous 100 years. Things were exceedingly quiet. With very few spots there was no significant magnetic activity on the Sun, and the solar wind dropped too. Its pressure fell by 22 per cent and less magnetic flux was being carried away from the Sun. The weak magnetic conditions on the Sun propagated into the Solar System.

  This reduction in the dynamic pressure of the solar wind could have an effect all the way out to the edge of the Solar System. The bubble of the heliosphere that swells out well beyond the planets is shaped and controlled by the forces of the solar wind and changes in the wind ripple outwards so that the heliosphere is perhaps smaller now than it has been in past cycles. This may very well have had the consequence that the Voyager 1 spacecraft on its mission to reach interstellar space didn’t have so far to go!

  The low levels of solar activity forced us to realize that the abundant sunspots and the wonderful explosive activity that we had watched and been fascinated by during the space age might not be the defining feature of our Sun. Some even speculated that the Sun might be headed for another Maunder minimum phase.

  Even though sunspot numbers were low, by 2009 enough had been seen for them to refine the cycle prediction methods. The legion of different approaches were amalgamated once more, and a revised prediction was issued by the Solar Cycle Prediction panel in May 2009. This time they suggested that solar maximum would occur in May 2013 but that cycle 24 would be smaller than average.

  As 2013 rolled along the sunspot numbers continued to disappoint, prompting a new speculation about solar maximum based on previous cycles that looked similar to cycle 24 during the rise phase. It seemed that sunspot maximum would now be delayed until late 2013, 2014 or maybe even 2015. But it turned out not to be quite that bad. Solar maximum occurred in late 2013 and continued into 2014. The hemispheres peaked at different times, which is not unusual. The rate at which sunspots appear in different hemispheres doesn’t have to be the same and this means that one hemisphere will reach solar maximum before the other.

  We now know that the Sun has been at its quietest since 1906. You could indeed say that it was a ‘rare’ drop in sunspot numbers, certainly on the scale of a human lifespan or the duration of the space age. Score one for the headline!

  But it may not be rare in the lifespan of the Sun. For me, it was fascinating to watch the Sun show a different side from the one I am familiar with, but the recent activity level goes to show that we shouldn’t judge the Sun on what we see during our relatively short careers. Why should such a short snapshot be a representative view of a star that has been living out its life for 4.6 billion years already? As cycle 23 waned the Sun entered a quiet phase that was deemed unusual by solar physicists who had been studying the Sun from the twentieth-century perspective. What can we learn about predicting the future from looking at the long-term activity?

  THE LONG VIEW

  Sami Solanki, a solar physicist and director of the Max Planck Institute for Solar System Science in Germany, introduced me to looking at the long-term solar activity: going further back in time than just the sunspot observations that were carried out in Europe. He was able to do this because there are ways to investigate the Sun’s long-term magnetic activity by using proxies rather than direct measurements. We met this before. The pulsing in strength and complexity of the Sun’s magnetic field is felt at the Earth by the level of shielding we get from galactic cosmic rays – high-energy charged particles.

  It’s a bit like how the Earth’s magnetic field protects us from the Sun’s harmful particles: the Sun’s magnetic field protects the entire Solar System from particles coming in from other stars. We are located inside the Sun’s atmosphere, and how many galactic cosmic rays can penetrate into the Solar System and reach us on the surface of the Earth is affected by the strength of the Sun’s magnetic field. Fewer are able reach us at times of solar maximum, when the Sun’s magnetic field is at its strongest.

  As we also saw in the previous chapter: high-energy cosmic ray particles leave their calling card through the production of the radioactive isotopes carbon-14 and beryllium-10. Radioactive particles are unstable and transform into other particles over time by breaking up. The half-life of carbon-14 is 5730 years, meaning that in that amount of time half of the particles in any collection will have broken up, and the half-life of beryllium-10 is 1.5 million years. Since these isotopes are created by cosmic ray impacts with oxygen and nitrogen in the stratosphere, and are ultimately stored in tree rings and laid down in layers of snow, they act as a time capsule through which we can study the past.

  Sami used these pieces of data to reconstruct the level of solar activity back over the past 11,400 years. Despite the uncertainties in the reconstruction of the solar activity from such data and the difficulties in trying to pick out an eleven-year cycle, long-term variations in the levels of these radioactive isotopes provide a way to look at the Sun’s activity over hundreds or even thousands of years, which in turn can be used to help understand what activity levels we might expect from the Sun in the future.

  These datasets show that the Sun went through other Maunder minimum phases before the European sunspot records began. This includes the Wolf and Spörer minima. The new perspective that this long-term view gives us shows that the Sun has grand minima which occur irregularly across the millennia. Over the last 9000 years of the history of the Sun sixty-six occasions of relatively high solar activity can be picked out of the long-term modulations in cycle size. They are somewhat arbitrarily defined but they are taken to represent times when the eleven-year solar cycles are large for several cycles in succession. These times of heightened activity have come to be called grand maxima.

  It seems that whilst judging the Sun on its eleven-year sunspot cycle is comfortable for human lifetimes, it is rather short-sighted on solar lifetimes. The complex ebb and flow of the sunspot numbers and the magnetic Sun have been taken for granted. But there are far longer-term trends that we are only beginning to appreciate. It seems that the state of the plasma flows and magnetic field movements evolve over timescales well beyond what we will ever be able to witness directly. We are probably only now just starting to gather the right information through our telescopes in space and on the ground. And we need to keep gathering these kinds of data for many generations to come. But with our current data we can still try to understand the solar rhythms that have an impact on our life here on Earth.

  Scientists are still trying to work out how each eleven-year sunspot cycle can affect the next one, particularly when the cycles become almost non-existent. It is fascinating that the solar cycle can apparently switch off for several decades so that no strong magnetic fields appear at the surface to form sunspots. But then, seemingly from nowhere, the Sun
recovers and the cycle re-establishes itself. The cycle must have been hidden the whole time, somewhere. This is what appeared to happen during the Maunder minimum. It’s a puzzle.What can we confidently say about the future?

  HOW ODD IS OUR CURRENT SUN?

  Well, Sami Solanki studied the long-term magnetic activity and found that although cycle 24 is weak, the levels of solar activity which we have witnessed over the last seventy years are exceptional. We now know that the last time the Sun was at such a high level of activity appears to have been over 2000 years ago. We might be trying to understand the Sun at a time when it is very much atypical.

  Sami suggests that during the last 11,400 years the Sun has only spent 10 per cent of the time being in such an active state. It seems we have been privy to a rare event. Our observations of the Sun during the space age, the time when solar physics rapidly progressed, seem to have coincided with the Sun being in a grand-maximum state. But now we appear to be coming out of the grand maximum of solar activity. This decline in overall solar activity appears to have been happening since around 1985.

  And, making a prediction from the cosmogenic isotopes, Mike Lockwood suggests that there is roughly a 10 per cent chance of the Sun going into a Maunder minimum in the next forty years and a 45 per cent chance within 150 years. If solar activity declines over the coming cycles then we may be looking at a dimmer Sun in the decades to come, but this is expected to have only a very small impact on global temperature. If the Sun goes into a grand minimum we could expect to see a cooling of 0.1–0.3 degrees Celsius as compared to the warming by 3 degrees expected from the rise in global temperatures that we are currently experiencing. But we may also see a larger fraction of relatively cold winters in Europe because of the effects of UV light on the stratosphere.