Site Section Links
KRONOS Vol V, No. 1
The Inconstant Sun
Copyright (C) 1980 by John and Mary Gribbin
The timescales that matter to man are those of decades, centuries, and, at the most, millennia. What happens over the next ten years is of vital importance to all of us; with a large global population and complex technological society, some plans now being made by governments and international agencies (construction of dams, roads, and other big projects) will have repercussions for several decades at least, affecting the lives of our children; and both historians and those concerned with the long-term future of mankind must be uncomfortably aware that the environment here on Earth, especially the climate, changes considerably on a timescale of a few centuries. Changes in the output of heat from the Sun, or in the nature of the radiation it emits, will have profound effects if they occur on any of these timescales. And, indeed, some climatologists now suspect that the climatic changes of the past thousand years or so can be explained by solar variations.
But now we are talking about the merest ripples on the surface of the Sun, a situation quite removed, it was thought before 1979, from anything going on in the deep interior. Then, in addition to Dr. Sakurai's astonishing discovery of a link between sunspots and solar neutrinos, Professor Robert Dicke, of Princeton University, came up with independent evidence that the behavior of the spots on the surface of the Sun is modulated by processes in the deep interior. As he puts it, there seems to be a chronometer hidden deep in the Sun. And all this becomes of crucial importance once we realize that those spots on the solar surface, along with other features of activity over the visible face of the Sun and out into space, vary on just the timescales that matter to man, over decades, centuries, and millennia.
What is a sunspot? In essence, it is simply a relatively cool region of the Sun's surface, appearing dark by comparison with the even hotter material around it – a mere 8,000 degrees or so instead of 9,000 degrees!* Individual spots may be anything from about 1,500 km across to 150,000 km, and they are associated with strong magnetic fields which may cool the region by inhibiting the all-important convection. Spots are simply the most visible manifestation of a whole range of solar activity – when there are more spots, the Sun is more active, producing a variety of phenomena related to solar magnetism, great flares which blast material up and out into space, disturbing the solar "wind" of particles which continually streams out past the orbits of the Sun's family of planets. The whole pattern of activity waxes and wanes over a cycle roughly 11 years long, with the solar magnetic field being reversed after each cycle so that it gets back to its starting state after 22 years. This magnetic cycle is generally regarded as the basic pattern of variation around which all the other bits and pieces, including changes in the number of sunspots, should be placed. But although modern techniques, including observations from satellites above the Earth's atmosphere, show that the whole of the surface activity of the Sun is undergoing this regular upheaval, with the changing number of spots simply a very minor symptom, like the rash on the skin of a measles victim, it is still common to call the pattern by its old name, the "sunspot cycle".
The basic 11-year sunspot cycle was only recognized about a hundred years ago, and ever since then it has been generally regarded as a very rough and ready "cycle". The spacing between years of peak activity has sometimes been as little as 7.3 years, and sometimes as great as 17.1 years, making a pretty vague "chronometer" and causing astronomers to puzzle over whether there is any "real" cycle or simply a succession of independent disturbances each of which take about 11 years to die down.
Professor Dicke's study completely transforms this view. He used all the power of modern statistical methods to ask what is in fact a very simple question: Do the changes in the length of individual cycles occur at random, or do they add up over the centuries to give a clearer pattern? And what he found is that, taking many cycles together and "averaging" them in the proper statistical sense, it seems that the sunspot numbers are being kept precisely in step by some great solar metronome. One example spells out how this works. Starting in 1761, the historical records* show that there were three very short cycles averaging 8.9 years, at the end of which the year of peak solar activity arrived 5.6 years "early" compared with the 33 year interval that "ought" to have occurred. But the next cycle was 17.1 years long, putting the sunspots exactly back in step with the 11-year rhythm, after an interval of 27 years and more. "It is as though," says Professor Dicke, "the Sun 'remembered' the correct phase for 27 years and then suddenly reset the sunspot cycle."
The story is the same right down the centuries, once it is seen with the powerful insight provided by statistical analysis of the records. Individual sunspot cycles may be long or short, but the average is kept very accurately in step.
There is only one reasonable explanation of such a phenomenon. It must be that something deep in the Sun is pulsing with a precise 11-year (or 22-year) period, and that all the surface effects of the solar cycle of activity are simply an imperfect reflection of this deep seated pulsation. The "errors" in the variation of the surface features must be due to imperfections in the way the "message" gets transmitted to the surface – it may take several years for the beats of this solar heartbeat to spread their influence outward, and sometimes it may take more years than at other times. It is far too soon yet for the theorists to have come to grips with this new evidence and to produce a better understanding of how the Sun works. That will take years, involving contributions from studies of oscillations of the solar surface, neutrino astronomy, and observations from orbit across the electromagnetic spectrum. But it is already further evidence of the variability of our own star, the Sun. And it provides appropriate background to look at the relationship between the solar cycle of activity and events here on Earth.
HOW VARIABLE IS THE SUN?
From the first Victorian astronomers who noticed the sunspot cycle down to the present day, people have tried to find links between the 11- and 22-year solar rhythms and the weather on Earth. Many claims have been made, involving claims of such patterns in rainfall figures for England, changing water levels in African lakes, and the number of times lightning strikes each year. Some extroverts have gone so far as to link the 11-year cycle with changes in stock market prices, but it has proved impossible to explain all of these claimed correlations in terms of a genuine solar-terrestrial link. The balance of evidence is that the link is real,* but just as the deep seated solar rhythm gets distorted by the time it reaches the surface, so the roughly cyclic changes in the surface activity of the Sun get distorted and diffused in their influence on the Earth. One hint of why this might be so comes from a combination of spacecraft observations and ground-based measurements.
For a hundred years, astronomers have been teased by the puzzle of whether the actual amount of heat being radiated by the Sun varies over the sunspot cycle – in other words, is a more active Sun hotter? Living at the bottom of the Earth's atmosphere, we simply can't make measurements of the "real" output of the Sun (the so called "solar constant") because an unknown amount of solar radiation is absorbed by the atmosphere before reaching our instruments. The ground-based measurements do suggest that the solar "constant" changes, by as much as I percent on a very short timescale linked with the lives of individual spots. But now instruments mounted on spacecraft such as Mariner 6 and Mariner 7 show no evidence of changes even one tenth as big as this due to the sunspots themselves, on the appropriate timescale of a few days or weeks. So something must be happening to the solar radiation as it passes through the Earth's atmosphere, and in particular it seems that since more ultraviolet radiation is produced when the Sun is active, the ozone concentration high in the stratosphere changes, altering the effectiveness of this layer as a barrier to the Sun's radiation.
All this tells us nothing about whether the solar "constant" changes by I percent or so from decade to decade, or between one century and the next – the measurements only looked at short-term effects of specific sunspots and associated activity. But it does hint very strongly that the atmosphere is a variable filter, sometimes letting more heat through to the ground than at other times. This alone is bound to produce an averaging effect, smearing out any solar cycles by the time their influence penetrates into weather systems. And this smearing effect is why it makes no sense to talk about changing patterns of weather and climate linked with changes in solar activity on any timescale less than a decade – roughly the average over a sunspot cycle.
If there is one burst of solar activity in one month of one year, the climatic patterns of the Earth won't be changed (although parts of our planet may feel the jolt). But if the Sun is either very active or very quiet for decades or centuries at a time, then we might expect to see the effects as broad, long-term changes in the climate of the Earth. One nagging piece of evidence provided the impetus for this possibility to be taken seriously. According to all the astronomical records we have, the seventeenth century was a period of greatly reduced solar activity, with very few spots at all seen for more than fifty years. This period exactly coincided with what climatologists know as the worst period of the "Little Ice Age," when winters were so harsh that many European rivers such as the Thames froze, glaciers advanced, and the repeated failure of crops brought recurring famines. Throughout the twentieth century the question has been asked: Was there really a dearth of sunspots during the Little Ice Age, or is it simply that nobody bothered to keep accurate records in the seventeenth century? In the mid-1970s the question was resolved by Dr. John Eddy, of the National Center for Atmospheric Research in Boulder, Colorado, who turned detective-historian to answer the riddle plaguing climatologists.
THE MAUNDER MINIMUM
Sunspots were known to the ancient Greeks, but this knowledge was lost in the West and the spottiness of the Sun only rediscovered by Galileo in the early seventeenth century. (The Chinese knew about sunspots throughout this period when no observations were being made in the West – but more of that later.) The realization that the Sun shows blemishes on its skin was among the "heresies" for which Galileo was persecuted; the central dogma of the Western Church in the centuries prior had been that the Sun was a perfect sphere, created in its perfection by God, so that suggesting imperfection in the Sun was seen in some quarters as implying that God could be fallible and was guilty of shoddy workmanship. Nevertheless, the facts couldn't be covered up once the rediscovery of sunspots was made, and indeed the religious arguments attending Galileo's observations should have ensured a keen interest in the Sun in the ensuing decades. Could the Church have suppressed written evidence of sunspot observations? Or could people have lost interest so quickly that no one bothered to publish news of such observations in the second half of the seventeenth century? Eddy's analysis of the historical record shows not; astronomers eagerly sought sunspots at this time and reported them at length when they appeared. They just didn't appear very often.
Today, even at the minimum of the sunspot cycle we might see half a dozen spots in the course of a year, while during the year of maximum solar activity in the cycle we might see 100 or more (such a peak of activity is due in 1980 or 1981). Since the seventeenth century, there has only been one year (1810) in which no sunspots were recorded; but after the realization that there is an underlying period to the sunspot cycle, Victorian astronomers of the 1890s
tried to find out how far back this period could be traced, and searched the records right back to Galileo's time. Pioneers Gustav Spörer and E. W. Maunder found that almost no sunspots were seen from about 1645 to 1715, a period dubbed the "Maunder Minimum" by modern astronomers. Between them, Spörer and Maunder published five detailed scientific papers on the subject in the years up to 1922, drawing on evidence such as the report in the Philosophical Transactions of the Royal Society in 1671, where the editor commented:
Cassini's own report of the phenomenon included the comment:
And the then Astronomer Royal, John Flamsteed, was so excited by the sighting of a spot in 1684 that he commented:
We can surely take it for granted that the Astonomer Royal was looking for sunspots in the intervening eight years! But, in this modern age, astronomers have been curiously reluctant to accept Spörer's and Maunder's accounts. There is still too much of an arrogant tendency to dismiss nineteenth-century scientists as bunglers, even though our present scientific knowledge builds from their work. So it was necessary for Eddy to silence those doubting Thomases by going back to the original sources, repeating and improving on the studies of his two pioneering predecessors, and publishing his results in the journal Science in 1976, a respectable journal and a modern date to bring the evidence to the attention of late-twentieth century science.
His research is impeccable and his results unquestionable. The Maunder Minimum is real; sunspot activity really did almost switch off between 1645 and 1715. Not only the sunspot records show this; when the Sun is more active the gusts of the solar wind spill charged particles into the Earth's magnetic field where, focused at the poles, they produce those great free light shows, the aurorae. More and brighter aurorae mean more solar activity; and significantly, very few bright auroral displays were seen in the second half of the seventeenth century. The solar wind also has another effect on the environment of our planet. When the Sun is active and the solar wind is strong, it shields us from cosmic rays from interstellar space. But when the Sun is quiet and the solar wind is weak, these cosmic rays penetrate the atmosphere in great quantities. There, they can react with nitrogen atoms to make atoms of the isotope carbon-14, the radiocarbon so useful in dating samples of old trees and other vegetation. Tree rings can be dated accurately simply by counting the layers of wood in a sample, and there are many trees around that are three or four centuries old. Each ring contains carbon (wood) which includes a proportion of carbon-14 depending on the age of the ring (which is known) and the strength of the cosmic rays in the year that the wood was growing. So tree rings give an accurate measure of the amount of cosmic radiation penetrating the solar wind each year – in other words, they tell us how active the Sun was each year. This very modern technique reveals the Maunder Minimum, with great clarity, as a period of increased atmospheric carbon-14 production, caused by a decrease in solar activity and in the strength of the solar wind.
So the Maunder Minimum is real and we have proof that the Sun can vary in a way which directly affects the Earth (certainly through carbon-14, perhaps by affecting the climate) for periods of tens or hundreds of years.*
With proof that the Sun is variable in an irregular way, and with the new tool of radiocarbon analysis of old wood samples as an aid, the past few years have seen astronomers delving back into the historical records, from both East and West, to find out just how much solar variation there was in the centuries before Galileo. The Western records provide circumstantial evidence from reports of aurorae, but the Eastern records include mention of sunspots themselves, as well as aurorae and other effects. The pattern that emerges is dramatic indeed for anyone who still suffers from the delusion that our Sun is either perfect, constant, or regular.
BACK TO THE BRONZE AGE
With the new carbon-14 tool proved as a good guide to solar activity, John Eddy has been able to use the technique to push back the story of solar variability to the Bronze Age. During the past 5,000 years, he finds, there have been times when the Sun was much more active than it is today, and also times when it was much less active. He describes the 11-year cycle as "but a ripple on an ocean of great and sweeping tides," tides which have brought at least twelve major "excursions" of solar activity away from the pattern we think of as normal since the Bronze Age. The carbon-14 record extends the study of solar variations back to 3000 B.C., about halfway between the present and the end of the most recent ice age here on Earth, and far beyond the scope of the written record.
Eddy uses the Maunder Minimum as his yardstick in assessing these events, defining it as an "excursion" of magnitude-l compared with the present day; the state of the Sun now, on this broad picture, is seen as moving toward a new maximum of activity, on a scale of a hundred years or more, the like of which has not been seen for many centuries, since the "Medieval Maximum" of the twelfth to fourteenth centuries. Using Eddy's names for the various excursions, and his scale in which the Maunder Minimum defines a unit of – 1, the twelve major events since 3000 B.C. are given in Table 1.
There is no evidence that these long-term changes in solar activity follow any cyclic pattern, and indeed clusters of positive excursions (more solar activity) and negative excursions (less solar activity) seem to be the rule, rather than a switch from positive to negative or vice versa. The two quiet periods of the millennium before Christ must have been dramatic indeed – each twice as extreme as the Maunder Minimum and together lasting for 360 [?] years, more than a third of that millennium. Like the Maunder Minimum period of the seventeenth century, these were cold centuries on Earth, further striking evidence that when the Sun is less active the Earth cools down. But on this sort of timescale, rather than the monthly or yearly flickers
Extreme periods of solar activity since 3000 B.C.
of individual bursts of sunspot activity, it seems that the temperature of the Sun itself may change. It's not that the temperature changes follow the sunspot cycle of activity; rather, the strength or weakness of the sunspot activity is a result of changes in the Sun associated with temperature changes. In 1976, Eddy described this viewpoint as a "hunch"; the discovery three years later of links between sunspot activity and deeper, buried processes inside the Sun must make the idea far more respectable than that. No wonder two generations of astronomers failed to detect evidence that sunspot changes cause changes in the temperature of the Sun – their investigations are described only too aptly as looking for a means by which the sunspot tail could wag the solar dog!
The peak range of variations in solar output over the past 5,000 years need only be about 1 percent to produce changes in temperature on Earth averaging 1 or 2 degrees, ample to account for all the climatic changes since the Bronze Age, but these leisurely changes over several decades would be very hard to measure directly. Even so, some measurements hint that solar output increased by about 0.25 percent in the first half of the twentieth century, a time of increasing solar activity when each sunspot cycle in turn was more active than the one before. But while the carbon-14 story makes a complete and convincing tale, such concepts are still so startling to astronomers brought up to believe in the dogma of solar constancy that it is well worth emphasizing that this is not the only record we have of changes in solar activity over the millennia.
Whereas Western philosophers from the time of Aristotle to the seventeenth century were handicapped by the degree of perfection and constancy in the Universe, Eastern scholars have always been much more interested in the changing aspects of the heavens, phenomena such as comets, new stars (novae and supernovae) and, of course the changing face of the Sun. They also had the right conditions for viewing sunspots without the aid of telescopes. As every astronomy primer stresses, it is dangerous to look at the Sun with the naked eye, let alone directly through a telescope (which is why sunspot observations today are made by projecting the telescopic image on a white screen), and you certainly won't see sunspots, even if they are there, by staring at the blazing Sun at noon. Some natural disturbance of the atmosphere is needed to dim the light from the Sun at the same time that spots large enough to be seen with the naked eye are around. Clouds are no good – they block out the Sun altogether. So what you need is haze in the atmosphere, or a dust storm, especially effective at sunrise or sunset when the Sun is low on the horizon and its radiation passes through a longer column of air than at noon.
These conditions occur much more frequently in the heart of a continent or, with prevailing winds from west to east, on the eastern edge of a continent where the winds are coming off the land rather than the sea. So England and western Europe are distinctly bad places for catching a naked eye glimpse of a sunspot, while China and Korea are ideal. Hardly surprising, then, that modern solar astronomers such as Dr. David Clark, of the Royal Greenwich Observatory, have turned to Oriental records for their studies of the changing Sun – and hardly surprising, too, that when those records do mention sunspots they very often comment at the same time that the brightness of the Sun was reduced. This does not imply that the Sun itself was dim, but that dust or haze in the atmosphere was obscuring its light and making it possible to pick out the spots against the glare of the solar disk.
These records go back in China in reliable form to the Han Dynasty (200 B.C. to A.D. 200), by which time a well-organized astronomical/astrological bureaucracy was established with the task of keeping the Emperor informed of portents from the heavens, and responsibility for maintaining the calendar. The historical records of each succeeding Chinese dynasty include a wealth of astronomical observations, and from the tenth century similar records are available from Korea. Generally, the astronomer/astrologers of the time regarded sunspots as a sign that something was up, but they were far from consistent in deciding just what such an omen portended; it is also quite possible that they didn't bother to mention all of the available "signs from heaven" in years when the Emperor seemed in no need of heavenly guidance. David Clark cites the example of the Chin Dynasty (A.D. 265 to 420) when very few "warnings from heaven" of any kind were published in the records just after the new Emperor came to power, when all seemed well with the world. Later in the dynasty, as dissatisfaction with the regime grew, a great many more heavenly events are put into the record as the then Emperor's celestial advisors struggled to find some comforting advice for him. So the astronomer/historian needs also to be aware of social and political changes over the centuries in order to make a reliable estimate of the times when there was no interest in the heavens, or when observations might have been deliberately suppressed. Partly for this reason, there is no immediate chance of getting any guidance to the 11-year cycle, if it existed, from the observations of 2,000 years ago. But there is very clear evidence of gaps in the record 100 to 200 years long, when very few sunspots were recorded even though there is no political reason why the evidence should be missing. These gaps, shown in Figure 1 which is based on Clark's work, fit in very nicely with the minima found by Eddy in his carbon-14 analysis. The Oriental records don't go back as far as the carbon-14 "record," and they are more patchy. But they confirm the accuracy of the carbon-14 method back to the Medieval Minimum of A.D. 640-710, a full thousand years before the start of the direct telescopic observations made in modern times. When all the evidence points the same way in the centuries where it overlaps, we can have a great deal of confidence in believing the one record, the carbon-14 measurements, that goes back furthest into the past. But there is still more to the story of ancient Oriental astronomical observations of the changing Sun and its influence on the Earth.
Figure 1: John Eddy's estimates of solar activity changes, determined from the carbon-14 record, are compared here with the sunspot sightings of Chinese and Korean astronomers. A major feature before the Maunder Minimum is now called the Spörer minimum; in addition to the major features of Table 1, this plot shows (at the top) two small dips in solar activity, the Medieval Minor Minimum around A.D. 1300, and the Little Maunder Minimum of the early nineteenth century. David Clark, who prepared the figure on which this illustration is based, has assessed the gaps in the Oriental record (A-E, middle plot) in terms of the political events of the various dynasties. Gap A occurred at a time of political upheaval in China, when many records were lost, so it may not represent a real decrease in solar activity. Gap B occurred after reunification, during a period when other records (including calendars) are very complete, and is probably a real change in solar activity. Gap C came at a time of low standards in astronomical work, just before the great reforms of the Sung Dynasty, and probably represents the incompetence of the observers rather than changes in the Sun – as the carbon-14 record confirms. The later gaps are confirmed by Korean records, as well as Chinese, so that even though gap D comes at a time when the Mongols were overrunning China we can be sure that it is real, while gap E closely corresponds with the Spörer and Maunder Minima known so well from other studies.
LINKING SUN AND EARTH
One astronomer, at least, has been bold enough to try to test for the presence of the 11-year cycle of activity over the entire period covered by the Chinese records. Dr. A. Wittman, of Göttingen University in Germany, has used the reasonable assumption that although it may be just about impossible to use these fragmentary records to search for a periodic variation if you don't know the "answer" in advance, it is a different kettle of fish if we know when the peak years of sunspot activity "ought" to have occurred in the past, and use the records to test whether more sunspots really were observed in those years than at other times. Again, you have to take account of political factors, and there is the risk of a circular argument since, of course, we know the period we hope to find. But Dr. Wittman finds that, putting the most modest possible interpretation on his analysis, fifty maximum sunspot years can be identified between 500 B.C. and A.D. 1600, when the telescopes took over. Of course there are gaps – centuries at a time when no peaks can be identified, just like the end of the seventeenth century. Some of these gaps are due to the Sun being quiet, some to lack of observations. But the key discovery is that all the sunspot peak years occur close to the dates expected simply by running the 11-year cycle backward into the past. As Dr. Wittman puts it, "It is highly likely that the sunspot cycle persisted without interruption throughout this time span. The mean period is equal to 11.135 years."
Today, with the evidence of Robert Dicke's work on the solar chronometer, we would rephrase this to say that throughout the past 2,500 years it seems that the clock inside the Sun has been running with a steady rhythm of 11 years. The sunspots themselves, when they are present, and any changes in solar activity or the solar "constant" which affect the Earth, are just the outward signs of more deep-seated changes in the Sun.
In the late 1970s, however, a new voice began to be heard again in scientific circles after years of silence during the upheavals of the previous years – the voice of China itself. With communication with the West reopened, news began to emerge of studies made by Chinese scientists in many areas, and one of the first to be reported was their latest work on the ancient records of solar activity and effects on the Earth. These reports appeared in journals such as Acta Astronomica Sinica, and still avoiding the "cult of the individual," the authorship of teams was identified as "The Ancient Sunspot Records Research Group" and the like.* Their analysis goes back to 43 B.C., providing a welcome independent check of the historical study made by Clark and others in the West. Like their Western colleagues, the modern Chinese find evidence in their records of Eddy's "Medieval Maximum" of A.D. 1100-1300, and gaps corresponding to the Spörer and the Maunder Minima. But the "Ancient Sunspot" group have gone further by devoting most of their analysis to an attempt to find other periodic variations in sunspot activity, way beyond the basic 11- and 22-year cycles.
Their results are rather unselective, by the standards of present day Western astronomy, and the team seems eager to seize on any trace of a cycle and claim it is real, trying to find a physical explanation. In the West, astronomers generally have to be hit over the head with a weight of evidence before they believe any of it, as poor Maunder himself found with the discovery of the seventeenth-century minimum. The best path, surely, lies somewhere between these extremes, so perhaps the reopening of scientific links between East and West will benefit both parties! Meanwhile, though, it is worth noting that the Chinese do seem to be persuaded that some influence of the giant planets of the Solar System affects sunspot activity.
Even if you prefer to file this idea under science fiction for the time being, there is no denying the opposite interaction, that changes in the activity of the Sun do affect the planets, and in particular that changes in solar activity affect the Earth. The very strong influence of the Sun on the Earth is shown most simply in new data from another Chinese group, reported in the same scientific journal as the "Ancient Sunspot" analysis.
This team has looked at changes in solar activity and changes in the rate at which the Earth spins since the early nineteenth century. There is no problem here of interpreting the evidence in the light of political history; happily we are back in the realm of modern astronomy with complete records available for all to see. Changes in the rate at which the Earth spins – the length of day – are known simply from timing the interval between each "midnight," when a chosen star is at its highest point in the sky (in principle this is easy; in practice it is a very painstaking operation, but we can leave the details aside). Since 1820, the length of the day has changed over a range of about eight milliseconds, leaving aside seasonal effects linked with changes in the atmospheric circulation and so on. Sometimes the length of day increases for a while, sometimes it decreases, in no easily predictable fashion (although some of the changes are linked with earthquake activity). The success of the Chinese team is that they have explained all the changes since 1820, and made a forecast of future changes up to the year 2000, simply by adding together twelve periodic effects, most of them related to known patterns of change in the Solar System, such as changing solar activity or changing alignments of the outer planets – or both.
This is an old idea, sparked off originally by the fact that the sunspot cycle itself is just over 11 years long, while the period of Jupiter in its orbit (the Jovian "year") is 11.86 years. The Chinese also find a lesser periodic ripple of 29.8 years, close to the orbital period of Saturn (29.5 years); a periodic ripple very close indeed to the interval between conjunctions of Jupiter and Saturn (alignments of the two planets together occur every 19.9 years, and a ripple exactly this length is found in the Chinese study); and the longest reliable period of all, 179 years, which corresponds to the interval between alignments of all the outer planets in the same configuration.
This last hint is the most tantalizing. Because of their movements at different speeds in their orbits at different distances from the Sun, the pattern made by the planets at any time seems to be ever-changing. But following the intractable laws of celestial mechanics, including both the influence of the Sun's gravity and little tugs from one another, the planets actually move so that this heavenly clockwork of the Solar System resets itself every 179 years. So the planets are aligned just the same way, as I write this chapter in 1979, as they were in 1800. Even with only 300 years of telescopic sunspot records, some Western astronomers had found a hint of a 179-year cycle in long-term solar activity; now the Chinese, with their study of the spinning Earth, seem to be confirming this.
INSERT KV3_68.TIF HERE
Figure 2: Changes in the length of day calculated from the cycles of outside influences, including sunspot activity and changes in the alignments of Jupiter and Saturn (dotted line), exactly match the actual measured changes since the early nineteenth century (solid line).
Changes in the Sun's activity do seem to affect the spin of the Earth – the length of day and changes in the alignments of large planets, such as Jupiter and to a lesser extent Saturn, do the same trick. This can be understood, perhaps, directly in terms of the gravitational (tidal) forces exerted by the giant planets on the spinning Earth. But there is also a temptation to speculate that, among the complexities being unraveled by this kind of study, the planets themselves – especially Jupiter – may affect the Sun, and the resulting solar disturbances then feed back their effects upon lesser planets such as the Earth.
It takes a bold scientist to include a 179-year periodic effect on the basis of a sample of data only 150 years long, but the "cycle" is needed to make the length-of-day curves fit exactly, and it does tie in with two physically known phenomena, the hint of a long period in sunspot variations and the realignment period of the planets. In both the solar activity variations and the changes in length of day, the cycle is far from dominant – if you like, it is a lesser ripple on the ripple of the 11-year variation. But if there is a clue here that all of the planets acting together do exert an influence on the Sun, no matter how small, then it makes more plausible the suggestion that Jupiter (by far the biggest of the planets, of course, bigger than all the rest put together) is indeed linked with the fundamental 11 year rhythm in some way. It may be hard to see how even a planet as big as Jupiter could do the trick, especially since the sunspot cycle is slightly different from the orbital period of Jupiter, and it might just be a coincidence that the solar clock keeps roughly the same time as Jupiter. But if the other patterns, involving Jupiter-Saturn conjunctions and alignments of all the planets, are also present, then the coincidence is stretched to the breaking point. This isn't yet respectable science, at least in the West. But I'm willing to stick my neck out in accepting the circumstantial evidence as compelling – I believe that in the very near future even Western astronomers will come to accept the influence of the planets on solar activity and will find out how the link works, probably with the aid of the next generation of spaceprobes and their instruments studying the Sun from above the murk of the Earth's atmosphere.
Speculation is allowable in science; indeed, it is often welcome as providing a signpost to the kind of further work that might be fruitful. Even a speculation that later turns out to be wrong is useful if it encourages people to find the right path. It is, however, vital that speculation not be presented as fact or dogmatic belief. The concept of the Sun as a perfect sphere would have been fine as a speculation which encouraged astronomers to look for imperfections, and then to discard the concept of perfection once sunspots were found. It caused trouble as an idea only because it became an article of faith, with established authority refusing to accept the new evidence. So I should be careful here to draw the line between established fact and speculative ideas, not only so that the respectable works don't get tarred with the brush of speculation, but also so that the vaguer ideas don't achieve an aura of belief that they don't yet deserve.
So – the work of Spörer and Maunder decades ago, backed up by the studies of John Eddy, David Clark, and Chinese astronomers in the 1970s, shows beyond any doubt that the Sun is a variable star in a timescale of decades, centuries, and millennia, with the visible sunspots best seen as a surface rash whose comings and goings hint at processes deep in the solar interior. It is possible, but by no means proven, that the development of this rash is linked in some way with disturbances from outside, dominated by the (presumably gravitational) influence of Jupiter. And it is certain that changes in the level of solar activity influence our planet, the Earth, in such a deep, fundamental way as changing the whole spin of our planet (by a few milliseconds). How else do the solar flickers affect the Earth – and can we make any forecasts of future terrestrial changes linked with those flickers?
The Book Case
[*!* Image]Available from Delacorte Press/Eleanor Friede
One Dag Hammarskjold Plaza, New York, NY 10017.