Climate sceptics who dispute the link between global warming and carbon dioxide emissions frequently argue that the increase in world temperatures over the past half century is part of a natural cycle. They cite previous periods in history when the climate has swayed into extremes, such as the "medieval warm period" when vines grew in northern England and the Vikings settled in Greenland.
Or they quote the Little Ice Age, which happened somewhere between the 14th and 18th centuries, when the Thames froze over and the Bruegels painted their snowy winter landscapes. History shows that climate is a variable feast, they argue, and what we are getting now is just another side dish.
At the heart of this view is the belief that natural variations in the Sun's activity are responsible for the warming of the past few decades. No one would dispute that the Sun is the main driver of climate – without it we would not have any climate. But what the sceptics are arguing is more subtle and complex.
They cite the work of two Danish scientists – Professor Eigil Friis-Christensen, director of the Danish National Space Centre, and Henrik Svensmark, who works in the same institute. Together, they have provided the rationale for believing that global warming has got more to do with natural variations in the cycle of sunspots on the solar surface than man-made emissions of CO2.
The theory is not that the intensity of the Sun has simply increased. Scientists are confident from 31 years of accurate, direct measurements of total solar radiation by satellites that there has been no overall increase in the amount of sunlight coming to Earth. Total solar irradiance, as it is called, has stayed remarkably constant and so cannot be held responsible for the warming of the past half century.
No, the theory of Friis-Christensen and Svensmark revolves around a far more subtle argument connected to the well-established 11-year cycle of sunspots that appear on the surface of the Sun. Sunspots are dark pools of magnetic activity that well up to the solar surface in periodic peaks of 11 years or so. When there are a lot of sunspots, the Sun is said to be more active.
In fact 11 years is only the average length of the activity cycle, which can vary between seven and 17 years. Shorter cycles of 10 years or less are associated with a more magnetically active Sun, when the solar wind of charged particles streams out towards the Earth with greater-than-normal intensity.
When sunspots are most active there is also a slight increase in solar intensity of 0.1 per cent. But this is hardly enough to account for the increase in global warming over the past half century, and this cyclical variation is not what Friis-Christensen and Svensmark are proposing as the cause of global warming.
The two Danes believe instead that there is a complex relationship between the length of the solar cycle and the amount of low-level cloud that forms in the Earth's atmosphere. Because shorter cycles are associated with a more magnetically active Sun, this affects the cloud cover and hence the climate on Earth.
The crux of their argument relies on several unproven suppositions. The main one is that clouds are more likely to form when solar activity is at its lowest and fewer magnetic pulses reach Earth. A second is that there will be enough of these clouds to reflect sunlight and lower global temperatures significantly, perhaps accounting for that famous Little Ice Age.
They also argue the opposite effect, involving yet more suppositions. When solar activity is at its height, the magnetic field and solar wind coming from the Sun is at its strongest – few scientists dispute this. However, they argue, this protects the Earth from the cosmic rays that emanate continually from deep space as a by-product of exploding stars.
Most controversially of all, they argue that cosmic rays are also responsible for creating charged particles in the air that become the seeds around which water condenses and clouds form. The effect of higher solar activity, so they claim, is to make the world less cloudy, and so warmer, by shielding the Earth from cosmic rays.
It is an elaborate idea and its sophistication is often lost on the many lay apologists of climate scepticism, including some who write for this newspaper. For them, it is just easier to cast doubt on the entire science of climate change, arguing that scientists disagree on the causes of global warming because some of them at least have evidence that it is a natural phenomenon connected with variations in sunspots.
But there is one lay observer who understands the theory better than most. His name is Nigel Calder, a veteran science writer who was editor of New Scientist in the early 1960s and is now a vocal climate sceptic.
Calder has also co-authored a book with Svensmark outlining why solar activity and not man-made CO2 is behind global warming. "Disdain for the Sun goes with a failure by the self-appointed greenhouse experts to keep up with inconvenient discoveries about how the solar variations control the climate," Calder wrote in 2007 when his book was published.
"[Svensmark] saw from compilations of weather satellite data that cloudiness varies according to how many atomic particles are coming in from exploding stars. More cosmic rays, more clouds. The Sun's magnetic field bats away many of the cosmic rays, and its intensification during the 20th century meant fewer cosmic rays, fewer clouds and a warmer world," he wrote.
So influential was Calder in promoting the solar activity theory of global warming that he was asked to give a seminar on the idea in the late 1990s at the European Organisation for Nuclear Research (Cern) in Geneva. Later, in 2000, he was acknowledged on the proposal document for a €1.5m experiment at Cern to test the idea that cosmic rays can generate the formation of clouds.
This proposal was co-authored by almost 60 leading scientists from some of the most prestigious laboratories across Europe. It cited from the outset the work of Friis-Christensen and Svensmark as the inspiration for the Cloud experiment, because of the apparent correlation they had found between global warming, solar activity, cosmic rays and cloud formation.
"If the link between cosmic rays and clouds is confirmed it implies global cloud cover has decreased during the last century. Simple estimates indicate that the consequent warming could be comparable to that presently attributed to greenhouse gases from the burning of fossil fuels," the Cern proposal states.
Most importantly, the Cern document reproduced two key graphs from previous scientific papers of Friis-Christensen and Svensmark. One showed an almost perfect correlation between global temperatures from 1860 to 1986 with the length of the sunspot cycle, taken from a Friis-Christensen study published in the journal Science in 1991. The other showed an equally dramatic correlation between cloud cover, solar activity and cosmic rays, taken from a joint Svensmark-Friis-Christensen study published in 1998.
"These observations suggest that solar variability may be linked to climate variability by a chain that involves the solar wind, cosmic rays and clouds. The weak link is the connection between cosmic rays and clouds," stated the Cloud proposal, which planned to test that weak link experimentally.
However, many scientists who have studied the work of Friis-Christensen and Svensmark point to a far more serious weak link. They argue that the two key graphs reproduced in the Cern proposal are based on flawed data. There is no correlation between global warming and solar activity, and no correlation between cloud cover and cosmic rays, the critics say.
The flaws were first identified by Peter Laut, a Danish scientist who was once science adviser to the Danish Energy Agency. Laut, now retired, demonstrated in a study first aired in 2000 and published in a peer-reviewed journal in 2003 that both graphs contained serious errors. When these flaws were corrected, the apparent correlations between global warming and solar activity, and cosmic rays and cloud cover, disappeared.
"Henrik Svensmark has never published a defence against my accusations in any scientific journal. Eigil Friis-Christensen has only once tried to refute my accusations," Laut said. "But, as I show in great detail in my 2003 article, his revised values are created artificially by introducing arithmetic errors. If you correct these, any support of global warming disappears.
"In spite of the non-existence of any proper defence, they both state publicly that they have rebutted my accusations many times. The proponents of solar influence upon global climate should be asked why they still cling to results that have been proven to be manipulated?"
Friis-Christensen and Svensmark deny that their data contain flaws, or have been manipulated, and both have questioned Laut's scientific credentials.
"One may wonder about the reason for Peter Laut's strong and unusually insulting wording in a scientific paper. However, it fits well with his methodology consisting of first writing false accusations, then totally neglecting the refutations, and finally referencing his very own claims as corroboration when publishing new accusations," Friis-Christensen said.
Asked why he has not replied to Laut's criticism, Svensmark replied: "I thought the comments so absurd, I didn't spend the time on doing that. Peter Laut is not a scientist. Of course he could have a point, but his only work in science is to make a critique of my papers and the papers of Eigil Friis-Christensen."
But other scientists working in the field have told The Independent that Laut's critique is correct and that the original papers published in 1991 and 1998 are seriously flawed. Six leading experts, including one Nobel laureate, agreed with Laut's analysis that the graphs of Friis-Christensen and Svensmark showing apparent correlations between global warming, sunspots and cosmic rays are deeply flawed.
Friis-Christensen now accepts that any correlation between sunspots and global warming that he may have identified in the 1991 study has since broken down. There is, he said, a clear "divergence" between the sunspots and global temperatures after 1986, which shows that the present warming period cannot be explained by solar activity alone.
Professor Jon Egill Kristjansson, a leading geoscientist at the University of Oslo, said that the divergence between global warming and solar cycles in the late 20th century "is now undisputed". He also points out that if Svensmark is right, there should have been a decrease in cosmic rays, but in fact over the past 50 years they have, if anything, slightly increased – despite statements to the contrary in the Cern proposal of 2000.
"Following Svensmark's mechanism, it seems that any cosmic ray explanation of current global warming can be ruled out," Egill Kristjansson said.
Although the Cloud experiment has only just begun and results are not expected for several years, Svensmark continues to promote the idea that cosmic rays play a significant role in cloud formation. However, other studies into cosmic rays, by both Egill Kristjansson and a separate team of Finnish scientists, have failed to find a link.
Egill Kristjansson said that there is still uncertainty about the "intriguing question" of cosmic rays but that does not mean everything else we know about global warming has to be thrown aside. "We should acknowledge that there is no doubt that the rapid 40 per cent increase in CO2 over pre-industrial levels that we have already experienced, and an expected continued increase, will inevitably lead to significant global warming, with possibly very dramatic consequences for life on Earth," he said.
So there is no need to invoke a complicated explanation for global warming involving disputed data on sunspots, cosmic rays and clouds, as some sceptics continue to do. The answer lies not in elaborate suppositions, but in the science and the data we can trust.