Claim: Fewer cosmic rays reach the lower atmosphere during solar maximum, producing fewer low clouds because they act as nucleation sites for cloud droplets, and fewer low clouds lead to higher average ground temperatures (Svensmark 2007). This is an amplifier for solar-caused warming of the Earth.

Why this Claim is Wrong: Unlike many claims, this one is not absurd on first glance. I can imagine having a sense of excitement on first considering this hypothesis and that it might possibly have something valid to say. It's worth a careful look. First, let's summarize the supposed mechanism, which is a bit more complex than some other claims: High solar activity means a stronger solar magnetic field permeating the solar system, which tends to deflect galactic cosmic rays (very high speed ionizing atomic nuclei) from hitting the Earth. Galactic cosmic rays (GCR's) create charged particles which are presumed to become cloud condensation nuclei (CCN's) from which water droplets form, especially in the lower atmosphere. Thus, the more GCR's, the more clouds and vice versa. Higher solar activity and hence fewer GCR's mean less clouds, reflecting less of the sunlight, and more is absorbed at the surface. In other words, the hypothesis claims - fewer GCR's -> fewer CCN's -> less low-level cloudiness -> more sunlight absorbed by the Earth's surface -> higher temperatures.

First, it is true that fewer cosmic ray secondaries reach the lower atmosphere during solar maximum. The reason is that the solar magnetic field is stronger near solar maximum, and therefore so is the shielding that it provides against cosmic rays from the Milky Way Galaxy in which we live (magnetic fields, as we learned in Astro 3 and Astro 4, cause a bending in the motion of charged particles).

What about the correlation of low clouds with cosmic ray flux? It is a complex series of steps needed to go from a cosmic ray hit in the lower atmosphere to a cloud condensation nucleus to a nucleated particle big enough to be called a cloud droplet and scatter sunlight and affect climate. The physics shows that there are already far more than plenty of nuclei available in the troposphere for cloud formation (dust, pollen, various aerosols). While there are still uncertainties in modeling the steps, note that Pierce and Adams (2008) find that the number of cloud condensation nuclei caused by cosmic rays modulated by the solar cycle is only about 0.1% of the total cloud condensation nuclei, which is “far too small to make noticeable changes in cloud properties based on either the decadal (solar cycle) or climatic time-scale changes in cosmic rays.”. A good summary of the state of aerosols and cosmic rays and cloud nucleation as of 2009 is here. Also, note that global night time temperatures are rising even faster than daytime temperatures. This fits well with CO2-induced IR absorption, but conflicts strongly with Svensmark's idea that it is cosmic-ray induced changes in albedo due to clouds.

So the theory of cosmic ray induced significant cloud cover looks - evening being maximally charitable - pretty weak. But what about the observations of clouds? After all, maybe some mechanism other than ones considered may be at work and yet induce a significant effect. A good way to see if GCR's affect cloudiness is to study so-called "Forbush Decreases (FD's) ", which are significant and rapid drops in the GCR intensity over a time scale of a few days. Since the lifetime of clouds is of order hours, studying cloud cover over these time scales will at least test if there is any demonstrable mechanism for GCR's to affect clouds, although not over the decades time scale relevant for climate change. Svensmark (2009), using only 5 FD's, says there is an effect from FD's, but Kristjansson et al. (2008) studied a larger sample of FD's and used MODIS satellite data, finding no correlation by several measures. NOAA (The National Oceanic and Atmospheric Administration) website summarizing global warming science, says "global analyses of cloud cover over land for the 1976-2003 period show little change". This site has a fairly comprehensive list of published papers (over two dozen) finding insignificant or no connection between cosmic rays and cloud cover.

More important, the ISCPP satellite data on global cloud cover (used in Svensmark's work) is known to be flawed (Evan et al. (2007) - authors and abstract given below...
"Arguments against a physical long-term trend in global ISCCP cloud amounts": Evan AT (Evan, Amato T.), Heidinger AK (Heidinger, Andrew K.), Vimont DJ (Vimont, Daniel J.), GEOPHYSICAL RESEARCH LETTERS Volume: 34 Issue: 4 Article Number: L04701 Published: FEB 17 2007
Abstract: The International Satellite Cloud Climatology Project (ISCCP) multi-decadal record of cloudiness exhibits a well-known global decrease in cloud amounts. This downward trend has recently been used to suggest widespread increases in surface solar heating, decreases in planetary albedo, and deficiencies in global climate models. Here we show that trends observed in the ISCCP data are satellite viewing geometry artifacts and are not related to physical changes in the atmosphere. Our results suggest that in its current form, the ISCCP data may not be appropriate for certain long-term global studies, especially those focused on trends.''

A good discussion of this viewing angle problem and how it's been used by the Svensmark camp is here. An article in EOS also shows how the claimed correlations are not actually present in the real data, here. The bottom line - claims of trends in cloud cover using this data set are not supported. So far, it appears ISCCP has not issued corrected data.

And finally, from a discussion on the RealClimate.org site..."Svensmark and others pointed to an apparent correlation between low-altitude cloud cover and cosmic rays. But after 1995, the beguiling fit of Svensmark's graph depends on a "correction" of satellite data, that the satellite scientists themselves say this is not justified. Here’s a quote…“I went to discuss this with the ISCCP scientists (downstairs) to get their opinion and this is a rough summary of the issues:
1) ISCCP inter-satellite calibration is done by offsets for individual satellites, not through trends. For instance, calibration is done using the warmest and coldest deciles from the satellite radiances. Whenever a new satellite stream is introduced there will be offsets in various parameters (so far ISCCP merges data from 33 different satellites) since the calibrations are done for a limited selection of targets, the satellites sometimes see different things and there may be subtle diurnal or angle related differences. Any new satellite can cause a jump, but it cannot produce a continuing trend. Calibrations are done directly against NOAA-9 (which was the best calibrated of all the satellites and serves as a gold standard for subsequent instruments). Thus the calibration of NOAA-14 (and NOAA 16 – from 2001) is performed directly against the NOAA-9 values, not with respect to overlapping satellites: http://isccp.giss.nasa.gov/JPEG/calib_be4after.jpg ). Thus there is no way the 1994 gap between NOAA-14 and NOAA-11 could have produced an ongoing trend in the ISCCP data.
2) The MS03 trend correction is based on the apparent difference between SSMI and ISCCP. But SSMI only sees water clouds and ISCCP sees both ice and water clouds, thus the trend could be a real effect, or it could be related to drifts in either or both instruments or to drifts in the calibration targets. It certainly can’t be assumed only to be a problem with ISCCP, and cannot have anything to do with the ‘gap’ mentioned above (since that would simply have produced an instantaneous offset)."
Bottom line: there is absolutely no good reason to put in a trend correction because of a gap in the satellites."

From reading the papers and responses (including Svensmark's here), my conclusion is that the cloud data itself is unfortunately not good enough to say much of anything about Svensmark's hypothesis, except that there is no support. There appear to be on-going problems with calibrations, with the inability of the ISCCP to differentiate between low, mid, and high clouds, view angle problems, and intra-satellite calibration problems. Since we're concerned with trends on the decadal time scale, it may be some time before this particular data is sorted out. While there probably is some correlation between cosmic rays and global temperatures, the simplest and best supported explanation is the already verified change in solar luminosity with solar cycle. Less radiation causes cooler temps, and this just happens to overlap with Svensmark's unrelated hypothesis, since more cosmic rays arrive when the solar cycle is a minimum. With the cloud observations muddled at this time, in judging Svensmark's idea we therefore put more emphasis on other observational/theoretical work, which shows clearly that cloud formation is far more sensitive to other factors (99.9%) than to ionized particle availability (0.1%) created by cosmic rays. I discussed this with a personal friend who is also a leading professor of atmospheric science and who develops models of cloud formation, and he agrees.

More important - since the cosmic ray induced cloud feedbacks clearly must happen (if they happen at all) on very short time scales (hours and days, not decades), and since there has been no secular trend in cosmic ray or solar activity since the mid 1950's (see the far right graph above), Svensmark's idea is a red herring as far as its relation to global warming. The data provide no support whatsoever for Svensmark's hypothesis as an explanation for global warming. In fact, cloud data show that far from acting to cool climate, clouds are reacting to global warming by amplifying it, i.e. acting with a positive feedback (Dessler et al. 2010, and described here)

It gets even worse for Svensmark. I recentlly found this nice summary of problems, and what's especially interesting is that 39 million years ago the Earth went through a magnetic reversal and for some time, the Earth's magnetic field was quite weak and let in far more cosmic rays. This event is recorded in paleo data as a spike in, for example, Beryllium 10, an isotope created by cosmic ray impacts on our atmosphere. Yet the temperature record shows nothing unusual happening during this period.

Here is an excellent in-depth analysis of the numerous flaws in Svensmark's work. Svensmark's response to the valid scientific criticisms of his work is also revealing - he claims a conspiracy is against him. The truth is, he's made strong claims that his work absolutely does not support. Svensmark teamed with writer Nigel Calder to publish a popular-level book "The Chilling Stars" on his hypothesis. The Institute of Physics has reviewed the Calder and Svensmark book here. From my own reading of interviews with Svensmark, I have to say I am inclined to agree with the Institute of Physics' unfavorable judgments. This 10 minute YouTube video "Climate Change - the Objections" is straight-arrow and shows good grasp on the science and gives a nice summary of the major problems with both Svensmark's hypothesis, and also those of Richard Lindzen.

While the early examinations of the Svensmark idea are negative enough, further work in the last few years has made the idea even less plausible. Here's a list of over a dozen papers published in quality journals which demonstrate flawed science, flawed judgment, and flawed interpretations of data which mar the cosmic ray hypothesis.

Update Aug 28, 2011.
The first paper showing results from the CLOUD experiment at the CERN particle accelerator, testing some ideas about cosmic rays' ability to generate clouds, has been published in Nature, by Kirkby et al.(2011). Climate denialists are making splashy claims that this work proves cosmic rays are causing global warming. But that's not at all the conclusion if you actually look at the paper and listen to the lead author himself, as you can do in this link and here. Instead, the CLOUD results only show that cosmic rays are a significant influence on the rate at which nanometer sized particles form, given the right chemistry. These are far too small to produce cloud droplets. Do these grow to become cloud condensation nuclei? There is no evidence that they do.

This video has some good visuals which are relevant for this claim

Update Nov 2013
The basic observations and physics I describe above show this cosmic ray hypothesis should not affect cloud cover, and new work confirms this. Sloan and Wolfendale (2013) find that no more than 10% of the warming seen in the entire 20th century can possibly be attributed to cosmic ray induced cloud changes. Erlykin et al. (2013) find no association between cosmic ray intensities and cloud formation, and that other work correlating large solar storms reaching Earth to clouds also show no effect. And finally, that even if cosmic rays did influence clouds, it should have produced global cooling, not warming, given observed cosmic ray trends. Fans of cosmic ray-induced climate change also believe that cosmic ray flux on Earth in the distant past, when the sun was in the inter-arm region of our Galaxy, could have induced large climate change. This idea has been refuted by Sloan and Wolfendale (in New Astronomy 2013), who use gamma ray astronomy and other astronomical data to show that variation in cosmic ray intensity between spiral arms to inter-arm periods is only about 10-20% over the past 1 billion years; not enough to support the climate or extinction claims. Also, Benestad (2013) finds there is no correlation between cosmic ray flux and either clouds, precipitation, or temperatures on Earth. And finally, Krissansen-Totten et al. (2013) use albedo measurements (MISR) from ISCCP (vs. the much less reliable cloud cover numbers from ISCCP) and correlates to cosmic ray flux over the past 13 years and find... "Our long-term analysis of MISR data finds no statistically significant correlations between cosmic rays and global albedo or globally averaged cloud height, and no evidence for any regional or lagged correlations. Moreover, epoch superposition analysis of Forbush decreases reveals no detectable albedo response to cosmic ray decreases, thereby placing an upper limit on the possible influence of cosmic ray variations on global albedo of 0.0029 per 5% decrease." More on these and other studies debunking the cosmic ray hypothesis can be found here

Return to Climate Denial Claims List

Return to Climate Science Main Page