Fiona MacDonald, ScienceAlert 8 Sep 09;
For as long as humans have looked up at the Sun, we’ve been wondering how solar variations affect the weather here on Earth. There has been a lot of research to suggest that events like sun spots and solar flares can cause the planet’s surface temperature to rise and fall, yet none of it has proved without doubt that there is a statistically significant or useful link. Thousands of years down the track, after many attempts to confirm the solar-weather relationship, a climate expert suggests it may be time to accept that any climatic effect of the small observed solar variations will be too minor to be of practical importance.
Barrie Pittock, one of Australia’s leading atmospheric scientists, has been sceptical about the link between solar variation and the Earth’s weather for a long time. He critically reviewed the research in 1978, 19831,2 and most recently, in Climatic Change on 28 August 2009.
Solar variations are events where the Sun’s energy changes. Sunspots are a common example, where due to intensified magnetic energy one patch of the Sun’s surface becomes cooler than its surroundings, causing a relatively dark spot. These occur in cycles of about 22 years – the amount of sunspots reaches a maximum and then declines for 11 years before reaching a minimum. When there are a large number of sunspots on the Sun it is suspected that Earth’s climate becomes slightly cooler. The important practical question is not whether there is a real effect but whether it would be large enough to be important for weather forecasting or in relation to observed climatic trends.
Far from completely writing off the solar-climate link, Pittock has in the past offered ground rules for improved testing that could validate future results. He said that the data was mostly unconvincing, of poor quality and often badly analysed. Some researchers even went as far as elaborating their hypotheses or selecting the data after looking at the results in order to explain or remove discrepancies, according to Pittock.
It’s not just Pittock who has his doubts; the Intergovernmental Panel on Climate Change (IPCC), filled with international atmospheric experts, could not find convincing proof that solar variations are a major cause of climate change on Earth.
Pittock concluded in the 1980s that “the evidence to hand suggests that if in the future more data and better analysis enable the detection of statistically significant relationships, these will account for so little of the variance [or variability] in the climatic record as to be of little practical value”.
Alvin M. Weinberg who wrote Science and Trans-science in 19723 said, “Science is at the one time both a purely intellectual quest for truth and understanding, and also the necessary prerequisite for practical applications and decision-making. Many questions can be asked of scientists, often with very practical applications, but which cannot be answered with the degree of certainty with which they would state a ‘scientific fact’ or regard a theory as in some real sense ‘proven’.”
Pittock believes this is the case with the solar-climate link, and that all we can do is use our judgement as to whether any claimed link is of practical importance. Another instance where this is the case is human-induced climate change. After all of the research, the many signs and An Incovenient Truth, Government’s and the public are starting to accept that we may be altering the Earth’s weather. Yet, this theory still has plenty of sceptics. Pittock says “Scientists cannot possibly predict accurately what will happen in the next 50 or 100 years, the best we can do is look at the limited data, theory and models and try to assess the probability of certain things occurring, such as rising sea levels, and the associated risks,” said Pittock.
But despite both human-induced climate change and the solar-climate link being ‘trans-science’ questions, there is a significant difference between the two problems, according to Pittock. He believes it’s time to step back and ask what we’re really trying to achieve by proving a link between solar variation and the Earth’s climate.
“If we assume this theory is true, we need to assess the benefit or risk that it presents. The changes in climate that supposedly result from solar variation aren’t large enough to cause global catastrophe, or even to provide useful climate predictions given other much larger variations. On the other hand, human-induced climate change is large enough to have possible consequences of an enormous magnitude.” Regardless of what we can prove as scientific fact, we need to act on the assumption that humans are rapidly causing climate change, because if we don’t we may face serious consequences, he says.
So what does the solar-weather connection sceptic think about human-induced climate change? “Quite early in my career I was wary of the idea. However, the growing body of evidence convinced me that my scepticism was unjustified and that I needed to get into the details regarding possible climate change in order to make this knowledge useful.”
Pittock admits that there probably are minor solar-weather connections, but he is convinced that if they were large enough to be useful for weather or climate forecasting, or to compete with the enhanced greenhouse effects as a cause of the presently observed climate changes, they would have been obvious and well-established by now.
In a paper recently published in the journal Climatic Change, Barrie Pittock sets out the main reasons why some researchers are still so interested in a possible link between solar variations and variations surface climate.
1. If the Earth’s climate is essentially driven by the Sun’s energy output, if it varies musn’t the Earth’s climate vary accordingly?
The trouble with this argument is that the Sun’s energy output varies very little. On the other hand the Earth’s surface climate has varied a lot over geological time. Natural climatic variability is due to all sorts of other “drivers” (causes) such as variations in the Earth’s orbit around the Sun, volcanic eruptions that put clouds of particles into the upper atmosphere, continental drift, and internal fluctuations such as natural quasi-cyclic events like El Nino, and processes that amplify small variations (termed “positive feedbacks”). It is therefore the relative magnitude of these various climate drivers and feedbacks that is critical as to whether solar fluctuations are important at the Earth’s surface.
Well established mechanisms are observable on short time-scales in the upper atmosphere where solar variations demonstrably drive changes in the ionosphere and even in the stratosphere, but these rapidly diminish with magnitude as we descend to the surface. In order to credibly claim that solar variations have noticeable effects at the surface, relevant positive feedback effects (which might magnify solar energy driven variations) must be postulated and proven via observations of intermediate effects and quantified energetics. Many have been suggested, but few if any have been reliably established.
2. What if solar variations can provide explanations for climate changes we don’t understand?
This is a tempting way to round off an otherwise quite valid set of observations with an interesting hypothesis. But is it justified or is it pure speculation? Here again, to be convincing a linking physical explanation or mechanism is needed, with quantification of the eventual effect. Otherwise, if it is published, it becomes a precedent in the literature which can be cited to justify another similar hypothesis grabbed out of the air. One unjustified hypothesis can easily lead to another being built on the first. This is common in the solar–weather relations literature.
3. It is easier to justify one’s disciplinary interest or specialty, e.g., solar or ionospheric physics, by suggesting that this research has practical application for weather forecasting. It may help get funding, and it might indeed prove to be valuable.
Such a reason is tempting but it is not rigorous science until mechanisms are established. But it can be better justified if others before you have made the same claim. What if there is something in it? Maybe it is worth pursuing just in case, but until there is convincing evidence one really ought to emphasise the speculative nature of the link. Several people making the same speculation does not make it correct.
William Burroughs, in his book Weather Cycles Real of Imaginary?4, discusses the question of the value of cycles in the weather as an aid in weather or climate forecasting. He states that these goals “more than justify the growing efforts to identify semblances of order in our climate.” However, he goes on to say “But this progress will be scant consolation to those enthusiasts who over the years confidently predicted that reliable patterns had been found and would enable forecasts to be made years or even decades in advance. In truth, almost everything they have studied so closely over the last few centuries is little more than the random noise of an immensely complicated physical system—full of sound and fury, signifying nothing.”
4. What if the Sun is causing global warming and not us?
This is one reason why the IPCC has had to devote a lot of effort to discussing solar variability. Maybe the advocates of solar causation might be right, so it must be taken seriously, as the IPCC has. The real problem here is that some self-styled sceptics use such suggested solar variability explanations, which are poorly established, as reason to dismiss, and to not act on the much better-established alternative.
Scepticism needs to be even-handed. If it is only applied to one side of an argument it is not objective and can be downright misleading. This is where rigour can help. Methodology and the quality of the data must be critically assessed, and statistics taken seriously. Could an eye-balled correlation be explained by chance? In order to deny that possibility we need to be sure of the dating and quality of the data, and understand any relationships between the data or effects of data smoothing that might affect the statistical significance of the result. Do we have a hypothesis that links the solar variations to the hypothesised climatic effect? And if so, can we document that it is operating? Any cyclic behaviour is unlikely to be established as statistically reliable, say at the 95 per cent probability level (i.e., less than a 1 in 20 chance of being accidental), unless there are at least four or five complete cycles of well-dated data. That is especially difficult to come by with proxy data such as pollen numbers or tree ring widths for cycles having periods of hundreds of years5.
It behoves authors, referees, editors and readers to bear these considerations in mind when dealing with claims in an area of science that has been hotly debated for decades. How much influence do solar variations have on surface weather and climate? And does the present claim, whatever it may be, bear close examination?
In summary, Pittock suggests that:
* any relationship between variations in solar output and the Earth’s surface climate is not yet shown conclusively to be other than by chance, and,
even if such a relationship were shown to occur it would be so small as to be of little practical use. If it were large enough to be useful It would have been well established by now. People have spent many decades looking for it.
Barrie Pittock is one of Australia’s top atmospheric researchers and authors. He is an honorary fellow of the CSIRO and author of “Climate Change: The Science, Impacts and Solutions” (CSIRO Pub., 2009).
References
1 Pittock AB (1978) A critical look at long-term sun–weather relationships. Rev Geophys Space Phys
16(3):400–420
2 Pittock AB (1983) Solar variability, weather and climate: an update. Q J R Meteorol Soc 109:23–55
3 Weinberg AM (1972) Science and trans-science. Minerva 10:209–222
4 Burroughs WJ (1992) Weather cycles real or imaginary? Cambridge University Press, Cambridge,
p 201
5 Wu J (2009) Possible solar forcing of 400-year wet-dry climate cycles in northwestern China. Climatic Change 96(4). doi:10.1007/s10584-009-9604-4
