Indirect Effects of the Sun on Earth’s Climate

By Mike Jonas – Re-Blogged From

And what might they be?” – Dr. Leif Svalgaard

For a long time, I have been bitterly disappointed at the blinkered lopsided attitude of the IPCC and of many climate scientists, by which they readily accepted spurious indirect effects from CO2-driven global warming (the “feedbacks”), yet found a range of excuses for ignoring the possibility that there might be any indirect effects from the sun. For example, in AR4 2.7.1 they say “empirical results since the TAR have strengthened the evidence for solar forcing of climate change” but there is nothing in the models for this, because there is “ongoing debate“, or it “remains ambiguous“, etc, etc.

In this article, I explore the scientific literature on possible solar indirect effects on climate, and suggest a reasonable way of looking at them. This should also answer Leif Svalgaard’s question, though it seems rather unlikely that he would be unaware of any of the material cited here. Certainly just about everything in this article has already appeared on WUWT; the aim here is to present it in a single article (sorry it’s so long). I provide some links to the works of people like Jasper Kirkby, Nir Shaviv and Nigel Calder. For those who have time, those works are worth reading in their entirety.

Table of Contents:

1. Henrik Svensmark

2. Correlation

3. Galactic Cosmic Rays

4. Ultra-Violet

5. The Non-Linear System

6. A Final Quirk



1. Henrik Svensmark

Back in 1997, when Henrik Svensmark and Eigil Friis-Christensen first floated their hypothesis on the effect of Galactic Cosmic Rays (GCRs) on Earth’s climate, it shook the world of climate science. But it was going to take a lot more than a shake to dislodge climate science’s autocrats. Their entrenched position was that climate was primarily driven by greenhouse gases, and that consequently man-made CO2 would be catastrophic (the CAGW hypothesis), and they were going to do whatever it took to protect their turf.

Those CAGW scientists were supported by remarkably little evidence. Laboratory experiments had verified the mechanics of CO2 as a greenhouse gas, but there was no empirical evidence that it was a major driver of climate. There were correlations, but inspection showed that temperature change always preceded CO2 change. The only support for CAGW came from climate models which had the assumed effects of CO2 built in. The models gave imaginary projections of what future climate would be like if CAGW was correct, but they could not reproduce past climate.

In 2003, Henrik Svensmark and Nigel Calder in the book The Chilling Stars [1] described how cloud cover changes caused by variations in cosmic rays are a major contributor to global temperature changes, and stated that human influences had been exaggerated.

Empirical evidence supported their theory, which they called Cosmoclimatology [2][3], and Henrik Svensmark had conducted an experiment to verify its mechanics. So Henrik Svensmark was fully justified in claiming that Cosmoclimatology “is already at least as secure, scientifically speaking, as the prevailing paradigm of forcing by variable greenhouse gases.”.

The next step was to publish in a peer-reviewed journal. Henrik Svensmark and his team at the Danish National Space Center (DNSC, now DTU Space) submitted a straightforward paper describing their experimental results to a peer-reviewed journal. They were stunned when the climate science tsars closed ranks and the paper was rejected. At this point, the clean-shaven Henrik Svensmark, as a kind of protest, decided not to shave until the paper was published. He had a pretty impressive beard by the time Experimental evidence for the role of ions in particle nucleation under atmospheric conditions [4] was eventually published in Proceedings of the Royal Society A. The process had taken 16 months.

Here we are, twenty years after the GCR hypothesis was first floated, and the CAGW paradigm is still in place and virtually unscathed. This is in spite of increasing evidence supporting Cosmoclimatology and in spite of the epic failure of climate models to predict climate. Paradigm protection has been seen many times in science, but I wonder whether it has ever been as corrupt and as extreme as it currently is in climate science.

I should have mentioned that there was strong opposition against experimental testing of Cosmoclimatology. Think about that – scientists trying to prevent a thoery being tested – and I think you will agree that my use of the word “corrupt” in the previous paragraph was justified.

2. Correlation

There is a strong correlation between solar activity and Earth’s climate. Jasper Kirkby wrote a wide-ranging paper, Cosmic Rays and Climate [5] in which he described the background to the planned CLOUD experiment at CERN, which would test the Cosmoclimatology theory.

In the paper, Jasper Kirkby presented a number of graphs which showed correlations between GCRs and climate. Of course, correlation is not causation, but as GCRs are controlled by solar activity the correlations do show a strong relationship between solar activity and Earth’s climate.

From the paper:

Over 500 million years:

Figure 1. Correlation of cosmic rays with temperature over the past 500 million years. [The paper’s Fig.11].

Note: The GCR flux varies as the solar system passes through the spiral arms of the Milky Way.

Over 12,000 years:


Figure 2. Correlation of GCR variability with ice-rafted debris events in the North Atlantic during the Holocene. [The paper’s Fig. 8].

The paper explains how the 14C and 10Be records are independent proxies for GCRs, and how ice-rafted debris relates to climate.

Over 3,000 years:


Figure 3. Correlation of δ18O and Δ14C with rainfall. [The paper’s Fig. 9].

The paper explains how Δ14C is a proxy for GCRs, and δ18O is a proxy for rainfall.

Over 2,000 years:

Figure 4. Correlation of GCRs with Central Alps temperature over the last two millenia. [The paper’s Fig. 3].

Over 1,000 years:


Figure 5. Correlation of GCRs with temperature over the last millenium, and also with glacial advances in Venezuela. [The paper’s Fig. 2].

The paper describes the underlying data.

In another paper, Beam Measurements of a CLOUD Chamber [6], Jasper Kirkby showed some 20th century correlations:


Figure 6. Correlation of GCRs with NH temperature. [The paper’s Fig. 12].

Figure 7. Correlation of sunspot cycle length with temperature. [The paper’s Fig. 6].

Solar cycle length probably has little to do with GCRs, but I included it here (a) to show that the sun’s effects might not be limited to just GCRs, and (b) to underline the fact that solar influence is harder to see on this timescale.

In total, the papers show that there is overwhelming empirical evidence that solar variation has a major effect on Earth’s climate on virtually all timescales from decades upwards. The main exceptions are the timescales on which the Milankovitch cycles dominate and make other influences very difficult to see. (Milankovitch cycles are caused by variations in Earth’s orbit, not by solar variations.).

Finally, Forbush Decreases provide an opportunity to test for solar impact over the very short term. A Forbush decrease is a rapid decrease in the observed galactic cosmic ray intensity following a coronal mass ejection (CME) (description from Wikipedia). Dragić et al [7] found a correlation between GCRs and Diurnal Temperature Range (DTR) during Forbush Decreases.

Figure 8. Observed DTR changes during Forbush Decreases (FD). Top panel is for FD intensity 7-10%, bottom panel for >10%. [Dragić paper’s Fig. 5].

There is typically an inverse relationship between DTR and cloud cover. NB. Although Dragić et al found correlation with GCRs, Laken et al [8] found that there was a “small, but statistically significant” influence from solar activity that was not caused by GCRs.

NB. Correlation of GCRs with climate do indicate that solar activity is involved, but not how. To link parts of climate to particular solar features such as GCRs or Ultra-Violet (UV) or solar wind or total irradiance, we will need mechanisms.

3.Galactic Cosmic Rays

The experiments that have been conducted on GCRs and Cosmoclimatology show some of the intricate complexities within Earth’s climate process. The journey of discovery was far from easy, with false starts, interacting factors, unanticipated problems, and, of course, a climate science establishment ready to throw up any obstacles they could.

In the end, Nigel Calder was able to claim that the whole chain of action from supernova remnants to variation in climate had been demonstrated, and that nearly all the breakthroughs had been made by Henrik Svensmark and the small team in Copenhagen.

The front end of the chain of action, from the stars to the solar modulation of cosmic rays, was well known. The rest of the chain, from there to Earth’s climate, had to be discovered and demonstrated.

3.1 The SKY Experiment

The 2006 SKY experiment at DNSC was aimed at testing the theory that GCRs could cause the formation of cloud condensation nuclei (CCN).

The background to the experiment is explained by Nir Shaviv in his article Cosmic Rays and Climate. After showing that empirical evidence for a cosmic-ray/cloud-cover link is abundant, he asks: However, is there a physical mechanism to explain it? In the SKY experiment, the DNSC team set up a cloud chamber to mimic the conditions in the atmosphere, in order to test for the physical mechanism. They then observed ionisation by gamma rays, and found that it did indeed lead to the formation of clusters of molecules of the kind that build cloud condensation nuclei.

This was the experimental result described in the much-delayed Royal Society paper referred to earlier [4]. As reported in the Royal Society’s press release: “Using a box of air in a Copenhagen lab, physicists trace the growth of clusters of molecules of the kind that build cloud condensation nuclei. These are specks of sulphuric acid on which cloud droplets form. High-energy particles driven through the laboratory ceiling by exploded stars far away in the Galaxy – the cosmic rays – liberate electrons in the air, which help the molecular clusters to form much faster than atmospheric scientists have predicted. That may explain the link proposed by members of the Danish team, between cosmic rays, cloudiness and climate change.”.

But there were a few more steps in the mechanism that still had to be tested.

3.2 The Link between the Sun, Cosmic Rays, Aerosols, and Liquid-Water Clouds

In 2009, Svensmark, Bondo and Svensmark [9] took a major step forward, when they used Forbush Decreases to demonstrate a complete link from cosmic rays through aerosols to liquid-water clouds.

The paper’s Conclusion begins: “Our results show global-scale evidence of conspicuous influences of solar variability on cloudiness and aerosols. Irrespective of the detailed mechanism, the loss of ions from the air during FDs reduces the cloud liquid water content over the oceans. So marked is the response to relatively small variations in the total ionization, we suspect that a large fraction of Earth’s clouds could be controlled by ionization.“.

But that phrase “Irrespective of the detailed mechanism” was a problem. They needed to know what the mechanism was.

3.3 The Aarhus Experiment

By 2006, the CLOUD experiment had been designed to test the mechanisms in the Large Hadron Collider (LHC) at CERN, a pre-experiment had been completed to check the validity of the main experiment, and by 2008 five new groups had joined the CLOUD collaboration [10], but the main experiment was taking a long time to get going. Opposition from mainstream climate scientists wasn’t exactly helping. So the DTU team decided to conduct their own experiment.

With help from Aarhus University, the team went back to the SKY cloud chamber, to conduct more advanced experiments, with the aim of demonstrating the complete mechanism by which GCRs create clouds.

The result was reported by Enghoff et al in their 2010 paper Aerosol nucleation induced by a high energy particle beam [11].

They reported: “We find a clear and significant contribution from ion induced nucleation and consider this to be an unambiguous observation of the ion-effect on aerosol nucleation using a particle beam under conditions not far from the Earth’s atmosphere. By comparison with ionization using a gamma source we further show that the nature of the ionizing particles is not important for the ion component of the nucleation.“.
3.4 The CLOUD Experiment

CERN’s CLOUD experiment reported its results in 2011. But shortly before that, the director-general of CERN made the extraordinary statement that the report would be politically correct about climate change. Nigel Calder explained it thus: “The implication was that they should on no account endorse the Danish heresy – Henrik Svensmark’s hypothesis that most of the global warming of the 20th Century can be explained by the reduction in cosmic rays due to livelier solar activity, resulting in less low cloud cover and warmer surface temperatures.“.

When the result was published in Nature [12] the next day, in Nigel Calder’s words it “clearly shows how cosmic rays promote the formation of clusters of molecules (“particles”) that in the real atmosphere can grow and seed clouds“.

Nigel Calder actually said rather more than that (read the full article). In particular: “[The new CLOUD paper is] so transparently favourable to what the Danes have said all along that I’m surprised the warmists’ house magazine Nature is able to publish it, even omitting the telltale graph.

Figure 9. The graph from the CLOUD paper.


A graph they’d prefer you not to notice. Tucked away near the end of online supplementary material, and omitted from the printed CLOUD paper in Nature, it clearly shows how cosmic rays promote the formation of clusters of molecules (“particles”) that in the real atmosphere can grow and seed clouds.

I can only suppose that leaving such an important graph out of the printed paper is what the CERN director-general meant by “politically correct”.

3.5 The Final Link

Needless to say, the climate science gatekeepers didn’t accept the findings. Their objection was that there was no explanation for the observation that sulphuric acid persisted at nighttime, whereas all the climate models assume that it cannot persist without ultra-violet light. (From Nigel Calder).

In 2012, Henrik Svensmark, Martin B. Enghoff and Jens Olaf Pepke Pedersen [13] published the final link in the saga. Their paper, Response of Cloud Condensation Nuclei (> 50 nm) to changes in ion-nucleation, found that ionisation from GCRs maintained the required sulphuric acid. GCRs continue unchanged at night-time, of course, while UV does not.

One final quote from Nigel Calder:

So Svensmark and the small team in Copenhagen have had nearly all of the breakthroughs to themselves. And the chain of experimental and observational evidence is now much more secure:


Supernova remnants → cosmic rays → solar modulation of cosmic rays → variations in cluster and sulphuric acid production → variation in cloud condensation nuclei → variation in low cloud formation → variation in climate.


Svensmark won’t comment publicly on the new paper until it’s accepted for publication. But I can report that, in conversation, he sounds like a man who has reached the end of a very long trek in defiance of continual opposition and mockery.“.

I hope to live long enough to see Henrik Svensmark receive the Nobel Prize for Physics.

Will climate science now recognise that it has been getting everything wrong for decades? I doubt it. Not until their leaders can be removed and replaced by scientists who will give as much critical scrutiny to CAGW as they do to competing theories.

4. Ultra-Violet

In the abstract for their 2007 book, Effects of the Solar Cycle on the Earth’s Atmosphere [14], Kamide and Chian explain that “the direct influence of the changes in the UV part of the solar spectrum (6 to 8% between solar maxima and minima) leads to more ozone and warming in the upper stratosphere (around 50 km) in solar maxima. This leads to changes in the vertical gradients and thus in the wind systems, which in turn lead to changes in the vertical propagation of the planetary waves that drive the global circulation. Therefore, the relatively weak, direct radiative forcing of the solar cycle in the stratosphere can lead to a large indirect dynamical response in the lower atmosphere.“. [I have not read the book].

In 2009, Gray et al [15], referring to improvements in SSI [Solar Spectral Irradiance] reconstructions, find a suggestion that “UV irradiance during the Maunder Minimum was lower by as much as a factor of 2 at and around the Ly‐a wavelength (121.6 nm) compared to recent S min periods and up to 5%–30% lower in the 150–300 nm region [Krivova and Solanki, 2005]. However, this work is still in its infancy.“.

The implication is that there could be at least two separate indirect solar effects on climate, namely GCRs and UV, and both might have played a role in the Maunder Minimum.

Gray et al also say “Interestingly, the large change observed by the SORCE SIM instrument was not reflected in TSI, the Mg ii index, F10.7, nor existing models of the UV variation. The implications are not yet clear, but these recent data open up the possibility that long‐term variability of the part of the UV spectrum relevant to ozone production is considerably larger in amplitude and has a different temporal variation compared with the commonly used proxy solar indices (Mg ii index, F10.7, sunspot number, etc.) and reconstructions.“. They add: “Most climate models [..] do not include the UV influence“.

Gray et al refer to GCRs too, but say that “The horizontal resolution of global climate models is tightly constrained by computing capacity since they must be global in nature and run for hundreds of years. Therefore, they do not resolve clouds explicitly, and inclusion of GCR mechanisms for assessment of their impacts requires careful parameterization“. In other words, climate models cannot include GCRs either.

If a climate model does not include GCRs or UV, is it really a climate model?

5. The Non-Linear System

Here’s a quote from a perhaps unlikely source, Christian Science Monitor: “In 1801, the eminent British astronomer [William Herschel] reported that when sunspots dotted the sun’s surface, grain prices fell. When sunspots waned, prices rose. With that, a 200-year hunt began for links between shifts in the sun’s output and changes in climate.

[..]”There are some empirical bits of evidence that show interesting relationships we don’t fully understand,” says Drew Shindell, a researcher at NASA’s Goddard Institute for Space Studies in New York. For example, he cites a 2001 study in which scientists looked at cloud cover over the United States from 1900 to 1987 and found that average cloud cover increased and decreased in step with the sun’s 11-year sunspot cycle.[..] From Herschel’s day through the early 20th century, scientists have offered correlations that “fall apart the longer you look at them,” he says“.

Faced with all the conflicting information and opinions, can we get a fuller understanding of them than Drew Shindell’s “fall apart“? I think we can.

There is one statement by the IPCC that should be displayed prominently in every climate scientist’s office: “The climate system is a coupled non-linear chaotic system, and therefore the long-term prediction of future climate states is not possible.” – IPCC TAR WG1, Working Group I: The Scientific Basis.

We are all so used to linear thinking that it’s difficult to go non-linear. But that is where we have to go.

In the climate context, “non-linear” means that the same influence (or input) can have different effects in different situations. For example, in certain conditions, the solar cycle might indeed affect the price of wheat or the US’s cloud cover for a time, but then as conditions change the effect will end. A corollary is that slightly different combinations of multiple inputs may have very different effects. An additional complication is that other influences may at times overwhelm the effects. Obviously, this makes everything a whole lot more difficult to analyse, but the idea that things “fall apart” comes from linear thinking. The truly serious problem is that it can be very difficult to distinguish between a real phenomenon that comes and goes, and a mirage. [By “mirage” I mean something that isn’t what it looks like.]. Let’s look at two of them. Are they real or mirage?

1. About the “pause” in global warming that had not been predicted by the models: “Near-zero and even negative trends are common for intervals of a decade or less in the simulations, due to the model’s internal climate variability. The simulations rule out (at the 95% level) zero trends for intervals of 15 yr or more, suggesting that an observed absence of warming of this duration is needed to create a discrepancy with the expected present-day warming rate.” – NOAA’s State of the Climate in 2008. When the “discrepancy” went past 15 years, the Met Office stretched the limit a little bit: “It is not uncommon in the simulations for these periods to last up to 15 years, but longer periods are unlikely.“. Ben Santer upped the limit to at least 17 years: “They find that tropospheric temperature records must be at least 17 years long to discriminate between internal climate noise and the signal of human-caused changes in the chemical composition of the atmosphere.” . The Met Office again: “several decades of data will be needed to assess the robustness of the projections”.

2. About the breakdown of the GCR-cloud correlation in the late 20th century: “Many empirical associations have been reported between globally averaged low-level cloud cover and cosmic ray fluxes. [..] In particular, the cosmic ray time series does not correspond to global total cloud cover after 1991 or to global low-level cloud cover after 1994 (Kristjánsson and Kristiansen, 2000; Sun and Bradley, 2002) without unproven de-trending (Usoskin et al., 2004).“. AR4 WG1 [Oct 2006]

Can you tell the difference between #1, a prediction that fails for 15 years or more but is not invalidated because there was climate noise, and #2, a correlation that fails for 15 years and is therefore invalidated in spite of there being climate noise? I thought not.

Here is a more reasonable way of looking at climate:

The sun influences Earth’s climate in various ways over various timescales. But these influences can be hard to detect at times because Earth has its own variations. Earth’s variations and the sun’s influences do not combine linearly.

Earth’s own variations include ocean ‘cycles’ like the AMO, PDO, ENSO and IOD, glaciers and ice-caps that come and go, and atmospheric shifts in the ITCZ and the Polar Vortex, to name but a very few. Man-made greenhouse gases are just a small player added to the mix (“Results suggest that from 1983-2009, cloud changes were responsible for a bit over 90% (90.6%) of global warming, man-made CO2 for less than 10% (9.4%).” – link).

When you see the correlations in section 2, you need to be aware of the timescale and the resolution. Those long timescales have poor resolution, so for example you can’t see a decade in a chart covering thousands of years. There would have been many short periods within each long period when conditions changed and the trend would break for a while. With that in mind, now look at the time when clouds broke with the GCR-driven pattern in the 1990s. Why wouldn’t they? It doesn’t alter the fact that a sun-cloud link has been firmly established. It just means that we have to keep our non-linear thinking cap on.

If we see a repeating pattern or a correlation in Earth’s climate, we can hypothesise about what caused it. If it subsequently disappears, we can’t then immediately dismiss it. In fact, until its mechanism has been firmly established and tested over time, we have to keep it under consideration and leave open the issue of whether it is real or mirage. Even when we have firmly established its mechanism, we still have to be open to the possibility that it will change under conditions that we haven’t anticipated.

The situation is made even more difficult by variable response times. For example, whenever heat is taken into the ocean, it may be any number of years before it re-emerges to influence climate.

In this very uncertain world of climate, one thing is just about certain: No bottom-up computer model will ever be able to predict climate. We learned above that there isn’t enough computer power now even to model GCRs, let alone all the other climate factors. But the issue of computer model ability goes way beyond that. In a complex non-linear system like climate, there are squillions of situations where the outcome is indeterminate. That’s because the same influence can give very different results in slightly different conditions. Because we can never predict the conditions accurately enough – in fact we can’t even know what all the conditions are right now – our bottom-up climate models can never ever predict the future. And the climate models that provide guidance to governments are all bottom-up.

6. A Final Quirk

The 100,000 year problem is a simple but striking example of how difficult Earth’s climate cycles are to interpret. The problem, as described, is that a 41,000-year cycle that had been regular for goodness knows how long suddenly changed to a 100,000-year cycle and stayed that way for the next million years, and no-one yet knows why.

But maybe even that 100,000-year cycle might be a mirage. If you look closely at it, you can see that it might actually be a 41,000-year cycle missing some beats.

Figure 10. Temperature and CO2 over the past 400,000 years, from Vostok ice cores. Temperature peaks are roughly 80,000 or 120,000 years apart, not 100,000.

How can such a strong cycle miss a beat? If Ellis and Palmer [16] are correct, then precession’s effect depends on conditions at the time. ie, it’s non-linear. And it seems that lack of CO2 is one of the conditions triggering the rapid temperature increases!

The science is settled? No way. This non-linear stuff is too much fun.



Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s