Climate Science Glossary

Term Lookup

Enter a term in the search box to find its definition.


Use the controls in the far right panel to increase or decrease the number of terms automatically displayed (or to completely turn that feature off).

Term Lookup


All IPCC definitions taken from Climate Change 2007: The Physical Science Basis. Working Group I Contribution to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change, Annex I, Glossary, pp. 941-954. Cambridge University Press.

Home Arguments Software Resources Comments The Consensus Project Translations About Support

Bluesky Facebook LinkedIn Mastodon MeWe

Twitter YouTube RSS Posts RSS Comments Email Subscribe

Climate's changed before
It's the sun
It's not bad
There is no consensus
It's cooling
Models are unreliable
Temp record is unreliable
Animals and plants can adapt
It hasn't warmed since 1998
Antarctica is gaining ice
View All Arguments...

New? Register here
Forgot your password?

Latest Posts


A detailed look at galactic cosmic rays

Posted on 27 September 2010 by dana1981

Henrik Svensmark has proposed that galactic cosmic rays (GCRs) could exert significant influence over global temperatures (Svensmark 1998).  The theory goes that the solar magnetic field deflects GCRs, which are capable of seeding cloud formation on Earth.  So if the solar magnetic field were to increase, fewer GCRs would reach Earth, seeding fewer low-level clouds, which are strongly reflective.  Thus an increased solar magnetic field can indirectly decrease the Earth's albedo (reflectivity), causing the planet to warm.  Therefore, in order for this theory to be plausible, all four of the following requirements must be true.

  1. Solar magnetic field must have a long-term positive trend.
  2. Galactic cosmic ray flux on Earth must have a long-term negative trend.
  3. Cosmic rays must successfully seed low-level clouds.
  4. Low-level cloud cover must have a long-term negative trend.

Fortunately we have empirical observations against which we can test these requirements.

Solar magnetic field

Solar magnetic field strength correlates strongly with other solar activity, such as solar irradiance and sunspot number. As is the case with these other solar attributes, solar magnetic field has not changed appreciably over the past three decades (Lockwood 2001).


Figure 1: Solar Magnetic Flux from 1967 to 2009 (Vieira and Solanki 2010)

Galactic Cosmic Ray Flux

Cosmic ray flux on Earth has been monitored since the mid-20th century, and has shown no significant trend over that period.

Figure 2: Cosmic Ray Intensity (blue) and Sunspot Number (green) from 1951 to 2006 (University of New Hampshire)

In fact cosmic ray flux has lagged behind the global temperature change since approximately 1970 (Krivova 2003).

"between 1970 and 1985 the cosmic ray flux, although still behaving similarly to the temperature, in fact lags it and cannot be the cause of its rise. Thus changes in the cosmic ray flux cannot be responsible for more than 15% of the temperature increase"

Figure 3: Reconstructed cosmic radiation (solid line before 1952) and directly observed cosmic radiation (solid line after 1952) compared to global temperature (dotted line). All curves have been smoothed by an 11 year running mean (Krivova 2003).

And since 1990, galactic cosmic ray flux on Earth has increased - "the opposite direction to that required to explain the observed rise in global mean temperatures" (Lockwood 2007).  In fact, cosmic ray flux on Earth recently reached record levels.  According to Richard Mewaldt of Caltech, "In 2009, cosmic ray intensities have increased 19% beyond anything we've seen in the past 50 years." 

Figure 4: Record cosmic ray flux observed in 2009 by the Advanced Composition Explorer (NASA)

Despite this record high GCR flux which we would expect to increase cloud cover and cause cooling, 2009 was tied for the second-hottest year on record, and the 12-month running mean global surface temperature record was broken 3 times in 2010 (NASA GISS).

GCR Cloud Seeding

In order for GCRs to successfully seed clouds, they must achieve the following three steps.

  1. GCRs must induce aerosol formation
  2. These newly-formed aerosols must grow sufficiently (through the condensation of gases in the atmosphere) to form cloud-condensation nuclei (CCN)
  3. The CCN must lead to increased cloud formation.

The first step is not controversial, and is being investigated by the CERN CLOUD experiment.  However, the second step is often glossed over by those espousing the GCR warming theory.  Freshly nucleated particles must grow by approximately a factor of 100,000 in mass before they can effectively scatter solar radiation or be activated into a cloud droplet (Verheggen 2009). Pierce and Adams (2009) investigated this second step by using a general circulation model with online aerosol microphysics in order to evaluate the growth rate of aerosols from changes in cosmic ray flux, and found that they are far too small to play a significant role in cloud formation or climate change.

"In our simulations, changes in CCN from changes in cosmic rays during a solar cycle are two orders of magnitude too small to account for the observed changes in cloud properties; consequently, we conclude that the hypothesized effect is too small to play a significant role in current climate change."

Numerous studies have also investigated the effectiveness of GCRs in cloud formation (the third step).  Kazil et al. (2006) found:

"the variation of ionization by galactic cosmic rays over the decadal solar cycle does not entail a response...that would explain observed variations in global cloud cover."

Sloan and Wolfendale (2008) found:

"we estimate that less than 23%, at the 95% confidence level, of the 11-year cycle changes in the globally averaged cloud cover observed in solar cycle 22 is due to the change in the rate of ionization from the solar modulation of cosmic rays."

Kristjansson et al. (2008) found:

"no statistically significant correlations were found between any of the four cloud parameters and GCR"

Calogovic et al. (2010) found:

"no response of global cloud cover to Forbush decreases at any altitude and latitude."

Kulmala et al. (2010) also found

"galactic cosmic rays appear to play a minor role for atmospheric aerosol formation events, and so for the connected aerosol-climate effects as well."

Although there was a correlation between GCRs and low-level cloud cover until about 1991, after that point the correlation broke down (Laut 2003) and cloud cover began to lag GCR trends by over 6 months, while cloud formation should occur within several days (Yu 2000).

Figure 5: Low cloud cover (blue line) versus cosmic ray intensity (red line) (Laut 2003).

Low-Level Cloud Cover

Unfortunately observational low-level cloud cover data is somewhat lacking and even yields contradictory results.  Norris et al. (2007) found

"Global mean time series of surface- and satellite-observed low-level and total cloud cover exhibit very large discrepancies, however, implying that artifacts exist in one or both data sets....The surface-observed low-level cloud cover time series averaged over the global ocean appears suspicious because it reports a very large 5%-sky-cover increase between 1952 and 1997. Unless low-level cloud albedo substantially decreased during this time period, the reduced solar absorption caused by the reported enhancement of cloud cover would have resulted in cooling of the climate system that is inconsistent with the observed temperature record."

So the jury is still out regarding whether or not there's a long-term trend in low-level cloud cover.

Inability to explain other observations

In addition to these multiple lines of empirical evidence which contradict the GCR warming theory, the galactic cosmic ray theory cannot easily explain a number of observed fingerprints of the increased greenhouse effect, such as the cooling of the upper atmosphere and greater warming at night than day.

Additionally, because cosmic radiation shows greater variation in high latitudes, we expect larger changes in cloud cover in polar regions if GCRs are successfully influencing cloud cover.  This is not observed.  Furthermore, examining the nuclear reactor accident at Chernobyl, ionization from the radioactivity would be expected to have produced an increase in cloud cover.  There is no evident increase in cloud cover following the accident (Sloan 2007).

Galactic cosmic rays can't explain global warming

In summary, studies have shown that GCRs exert a minor influence over low-level cloud cover, solar magnetic field has not increased in recent decades, nor has GCR flux on Earth decreased.  In fact, if GCRs did have a significant impact on global temperatures, they would have had a cooling effect over the past 20 years.

This post is the Advanced version (written by Dana Nuccitelli [dana1981]) of the skeptic argument "it's cosmic rays".

0 0

Printable Version  |  Link to this page


Comments 1 to 27:

  1. Thank you, very informative and well written! There is one paper you could add, Rohs et al. 2010: A correlation study of high-altitude and midaltitude clouds and galactic cosmic rays by MIPAS-Envisat, They also examined forbush events and they found a small effect (while Calogovic et al. 2010 found no effect), but in the wrong direction!
    0 0
  2. Good article Dana. One variation I’ve seen on this skeptic argument is that the length of the solar cycle (10.66 years on average) is more meaningful to cloud effects and/or temperatures than the amplitude of the cycle. Cycle 23 was a long one, ~12.5 years long, which means we should experience cooling over the 8 years following the 2002 peak. Of course they predict the next few cycles to be long ones too. I’m sure it’s no more than astrology but some clear evidence against this line of reasoning would be nice to have. A snippet of this idea is presented below with a link to all the gory detail. “The cycle length of cycle 22 which peaked in 1990 was 9.8 years. Landscheidt has suggested a lag of up to 8 years between solar peaks or troughs and temperatures, which would mean a peak warmth from 1995 to 1998. Global temperatures appear to have peaked in 1998. The current longer quieter cycle 23 may be behind the cooling in the last 7+ years.” – Ultralong Solar Cycle 23 and Possible Consequences May 26, 2008
    0 0
  3. Or is a peak in warming 8 years following the 2002 peak? Their line of reasoning seems inconsistent.
    0 0
  4. For once I have to agree with your conclusions. Galactic Cosmic Rays do appear to have an effect on cloud formation but it is not the dominant effect that Svensmark, Friis-Christensen and Shariv might wish for. However, the correlation between "Cosmic Ray Intensity" and "Sunspot Number" shown in your Figure 2 is striking. Likewise, Figure 4 that shows the imperfect correlation between "Cloud Formation" and "Cosmic Ray" variation. Personally, I don't buy the idea that GCRs have a major impact on climate but these are interesting correlations that need to be better understood.
    0 0
  5. The GCR-cloud seeding link isn't so obvious: A few comments on the CERN CLOUD experiment here and here. Note two key NASA reports during the last solar minimum: Cosmic rays hit space age high Solar wind loses power The Pierre Auger GCR observatory has as yet unpublished indications of increased frequency of GCR events as the solar magnetic field wound down. All point towards ideal conditions for GCR-induced cloud formation and the cooling that is supposed to accompany it. Annual reports from the US of that so-called cooling: Based on data from January through December [2009], the average annual temperature for the contiguous U.S. was 53.1 degrees F (11.7 degrees C), which is 0.3 degrees F (0.2 degrees C) above the 20th Century average Based on data through the end of 2008, the contiguous U.S. experienced a nationally averaged temperature that was the coolest in more than ten years. The average temperature of 53.0°F (11.7°C) was 0.2°F (0.1°C) above the 20th century (1901-2000) mean. Conclusion? A definite maybe. This particular piece of science isn't settled.
    0 0
  6. muoncounter - note that increased GCR flux would mean increased cloudcover, increased albedo, and global cooling, if the GCR theory were correct. Quite obviously we're not currently experiencing cooling, as 2010 has been a very hot year. I may update the article to include some of that information though. I recalled reading the NASA reports but couldn't remember where to find them. Thanks for that.
    0 0
  7. #6: "Quite obviously we're not currently experiencing cooling" Yes, all the GCR cooling thus far has definitely been underwhelming.
    0 0
  8. Dana says "In addition to these multiple lines of empirical evidence which contradict the GCR warming theory, the galactic cosmic ray theory cannot easily explain a number of observed fingerprints of the increased greenhouse effect, such as the cooling of the upper atmosphere" Unrelated to the whole cosmic ray hypothesis, the types of radiation are more variable through the solar cycle than TSI, most notably UV. This in itself could most certainly cause upper atmosphere cooling. Through its effect on ozone production and obviously heating o the stratosphere, is primarily through uv absorption by O3. This in itself could most certainly have an effect on the jet stream, and the pressure systems in the troposphere.... there is a lot to be learnt as far as solar effects on climate go.
    0 0
  9. Dana, thanks for mentioning in your review that the main problem about the GCR hypothesis is particle growth. That physical part of it has no solution to date. I do not believe that any significant effect on cloud nucleation exists from GCR at all. No correlation has been demonstrated to the necessary level of confidence.
    0 0
  10. Dana Good Post Personal view is that Svensmark has at best possibly identified a small contributing factor to the residual climate varibility left after allowing for AGW, El Nino's etc. Useful line of marginal investigation (and he may be right at that level) but don't stop the presses. And this post supports that. However one area I would pick up on, just to be pedantic.... Links to Chernobyl. The theory of GCR influences is specific about the energies of the radiation involved, height in the atmpsphere etc. The suggestion that Chernobyl might be expected to provide supporting evidence is drawing a long bow.... The difference between Alpha, Beta & Gamma radiation etc, and the energies envolved. I personally wouldn't expect Chernobyl to provide any evidence pro or con.
    0 0
  11. Phillipe Chantreau @9 Good Comment The problems with Svensmark's theory seem to be: 1. Particle growth. 2. Competition with other sources of CCN's that allow his Cosmogenic sourced CCN's to predominate and thus drive the impact. 3. Questions re the total impact of CCN's on the net positive or negative impact of the radiative forcing impacts of Cloud changes. 4. No evidence of long term trend in Cosmic ray fluxes to support long term trend in climate influences. So.... A minor contributing factor to residual climate variability.. Yeh, Maybe. A major driver of climate change. Show me the data. Otherwise he's dreaming. And his act of climbing into bed with the more radical climnate denialist cause has damaged his case. He may well have identified a modest contributor to residual climate variability. But by overselling his case and getting into bed with the whacko's, he has done himself a dis-service. Sad really.
    0 0
  12. #10:"I personally wouldn't expect Chernobyl to provide any evidence pro or con." I don't see how low energy alpha or beta radiation could go far enough to ionize any part of the atmosphere. To consider gamma radiation as a factor, you could look at Gamma ray bursts. In March 2003, a large GRB was detected: The burst poured out a thousand trillion, trillion times the gamma rays seen in a solar flare. When measured more than one hour after the burst, the afterglow was still about as bright as a 12th magnitude star. But its not clear at all if this radiation seeds clouds.
    0 0
  13. I took the Chernobyl discussion straight out of Sloan and Wolfendale (linked in the article). "We estimate that the increase in ionization from this [Chernobyl] radioactivity relative to that produced by CR is a factor of ~15 in the immediate vicinity of Chernobyl (50-52.5◦ N, 30-32.5◦ E) and a factor ~3 in the fallout region..."
    0 0
  14. The basic problem is that cosmic ray seeded CCNs are not the only aerosols that can seed clouds.
    0 0
  15. "Fortunately we have empirical observations against which we can test these requirements." Somehow I stopped to think about the connotation of this sentence from the prolog. Question: Why is the word 'fortunately' appropriate in this context? (a) Because it is such a joy to be able to 'debunk' all skeptic arguments? (b) Because it would be terrible if somebody holding opinions other than ours would ever be right, plus it would be so embarassing if main stream climate scientists were ever wrong?? (c) Because if our climate was governed by outer space, there would be nothing we could do about it - now, all we have to do is stop burning coal and oil? (d) Because of other reason (please specify!) (Check one)
    0 0
  16. No Argus. It's fortunate that we live in the times that we do with all the advantages of scientific instruments unimaginable to earlier generations. It's worth reminding ourselves of that from time to time. It's not just dentistry and antibiotics that make this a good time for us. There's a whole infrastructure of scientific endeavour which we don't bother to look at most of the time. When we need it, it's there.
    0 0
  17. Argus (d) Fortunately rather than being ignorant and subject to argument by arbitrary assertion, we can use facts to check our understanding of the world.
    0 0
  18. Argus, it's *always* fortunate when there is empirical data against which we can test *any* theory. If his theory were correct, Svensmark would be fortunate that the data exists to confirm it.
    0 0
  19. Interesting article in this weeks New Scientist about possible links between the Solar Cycle and climate. The mention Svensmarks theory and that it has issues, but also mention another mechanism where GCR's may have more impact. Rather than creating CCN's to seed cloud formation, they may contribute to explaining the fact that clouds are often charged on their top and bottom surfaces as GCR's may ionise existing droplets. Speculation is that this could impact on the lifetime of a cloud, precipitation rates from a cloud etc. Certainly a more plausible then Svensmarks more convoluted mechanism. But these may not be the only impacts of Solar variability. Possibly variations in solar wind and magnetic field strengths could impact on the rate of deposition of cometary and meteoritic dust from space, another source of CCN's. Preliminary evidence suggests that although TSI only varies by about +/- 0.1% over a solar cycle, the Ultraviolet component of this may vary by as much as 1-2%. Ultraviolet is preferentially absorded in the stratosphere by ozone, but in addition UV is also part of the cycle vy which Ozone is actually created and destroyed, as well as the processes by which CFC's are destroyed, all of which can impact on climate. Also the stratosphere is where methane is converted to CO2 & Water so changes in the chemistry up there due to Solar cycle influences could impact on levels of Methane. Interestingly, Methane levels in the atmosphere plateaued for much of the 2000's, only to resume rising towards the end of the decade, roughly in line with the 'It hasn't warmed since 1998' period. Coincidence? And there may well be other mechanisms yet to be found by which different components of the solar system that vary with the solar cycle may have climate impacts - UV, Solar wind, Magnetic Field strengths, Simply correlating just GCR's to Atmospheric Temperature variability alone and implying a single major forcing here is stretching too long a bow. I suspect we will find a range of different mechanisms, caused by different phenomena associated with the solar cycle that each make modest contributions to climate variability. This doesn't in any way detract from the central theories about AGW that are based on solid radiative physics. Rather these would be simply additional secondary mechanisms that contribute to climate variability. Put simply, AGW currently describes the physics driving underlying trends. A range of other factors, including but not limited too solar cycle factors probably explain shorter term climate variability that is overlaid on top of the underlying trend. And periods such as a decade of lesser warming is still just short term variability.. Never forget, to produce his graph correlating GCR's with temperature Svensmark not only removed the impacts of ENSO, Volcanoes etc to reveal the residual impact, but also he had to remove a .14 DeC PER DECADE TREND as well. AGW due to GH Gases AS WELL as Solar Cycles influences. Not INSTEAD OF.
    0 0
  20. #19: "they may contribute to explaining the fact that clouds are often charged on their top and bottom surfaces as GCR's may ionise existing droplets." Glenn, Interesting post. Looking at the abstract of the GRL article cited in the New Scientist post you mention, I don't see any specific mention of GCRs as opposed to solar cosmic rays: Cloud edge droplet charging is expected from vertical flow of cosmic ray generated atmospheric ions in the global electric circuit. Its long been known that solar cosmic rays (mostly muons - my personal favorite), are ionizing. The origin of cosmic ray research was an effort to explain why charged, shielded electroscopes spontaneously lost their charge. Solar cosmic rays are vastly more abundant than GCRs and thus would be more likely candidates for 'cloud edge droplet charging.'
    0 0
  21. muoncounter @20 I wasn't aware of the distinction between the two sources of Cosmic Rays. Do you know whether the underlying fluxes of SCR's from the Sun are constant; as distinct from the levels reaching the Earth which are influenced by Solar cycle influenced shielding of the planet? Or do the raw fluxes vary with the Solar cycle?
    0 0
  22. #21: Solar cosmic rays are produced by solar wind ions (protons mainly) striking gas nuclei in the upper atmosphere. The flux to the upper atmosphere is thus a product of solar activity (CMEs and flares), observed for several years now by NASA's ACE satellite. The flux at the surface, measured for many years by an international network of neutron monitors and more recently by a network of 'muon counters', varies on a scale of hours, days, months for different types of solar events. These are typically low energy particles, moving at relativistic speeds. The earth's magnetic field (as distinct from the solar or interplanetary magnetic field) is a strong modulator of solar cosmic ray flux. That's why auroral displays (the interaction of charged particles in the earth's magnetic field) are mainly visible in the high latitudes. GCR flux, even in times of the lowest solar magnetic field intensity, is much smaller than solar cosmic ray flux. The key point about GCRs is that they are much higher energy than solar cosmic ray particles.
    0 0
  23. On the issue of GCR Cloud Seeding You seem to have completely ignored the body of work by Harrison and others who have made in situ cloud observations which "are consistent with enhanced production of large cloud droplets from charging at layer cloud edges." There's an example of their work below but also many more from what I can see in literature searches. Harrison and Maarten 2009
    0 0
  24. #23: The Harrison and Ambaum 2009 paper you cite is about earth's electric field; it has nothing to do with this topic. The words 'cosmic ray' appear in the abstract, keywords and a non-specific reference to a 1989 paper. The statement in the abstract ... arising from the combination of distant thunderstorms, Earth’s conducting surface, a charged ionosphere and cosmic ray ionization is utterly inconclusive. Neither 'GCR' or 'galactic' appears at all.
    0 0
  25. Responding to Muoncounter: ""may be a connection between low solar activity and lack of major volcanoes, ... may also be an impact of galactic cosmic rays, ... on cloud albedo." They're way off topic, but those are new ones on me. If you can find an appropriate thread, care to explain?" This is just a speculative thought of mine about Svenmark's hypothesis. My first thought when I heard Svenmark's hypothesis is exactly described by El @14 above. There is an abundance of Cloud Condensation Nuclei, particularly over ocean (salt from sea spray), desert (dust) and forest (aramotic compounds released from leaves). Adding one more source of CCN is therefore unlikely to increase cloud cover. However, adding additional CCN is known to reduce the average droplet size within clouds. This can be seen most easily in ship tracks: It is also a known effect of industrial polution, with clouds downwind of cities typically having smaller than usual droplet size. (This does not get much discussion when deniers discuss the Urban Heat Island effect, and surface station placement, for reasons that will become obvious.) Decreasing cloud droplet size has two effects. It decreases the probability of rain, and it increases the albedo of the cloud. The former may result in longer lasting clouds, but if humidity drops, it is unlikely to compensate for the faster evaporation due to the increased surface area per unit volume from smaller droplet size. However, the increased albedo is real, detectable, and will have a cooling effect. So, assuming that Svenmaark is right in claiming that GCMs form CCN, and he has at least mounted a plausible case, then increased GCMs will increase cloud albedo by a small amount. It is obviously not a large amount because, if it were, the effect would be very obvious in the temperature record as a strong correlation between temperature and Sun Spot Number. But it may explain the small correlation seen between temperature and SSN surface temperatures, and some of the correlation between deep solar minima and cool temperatures which far excede the ability of TSI changes to explain.
    0 0
  26. Tom, "increased GCMs will increase cloud albedo by a small amount. It is obviously not a large amount because, if it were, the effect would be very obvious in the temperature record " Thanks. But that albedo increase may well be below any reliably detectable threshold. It would seem to be a second-order mechanism to the whole Svensmark idea, which has thus far eluded reliable detection on its own. It was really the low solar-lack of volcano activity link that made me wonder what I had missed way back when in my 'shake and bake' class. Editors: an unclosed italics tag seems to have infected this thread, immediately after Fig 4.
    0 0
    Moderator Response: [Daniel Bailey] I took a look at it; the error is in the html editing window. It's like The Matrix, raw code. Gonna take a bigger boat than I have to fix...
  27. DB, (or should I call you Spoon Boy?) Do not try to bend the spoon fix the italics tag — that's impossible. Instead, only try to realize the truth: there is no spoon italics tag.
    0 0
    Moderator Response: [DB] Yes, Obi-Wan; this Padawan realized his error, adapted, improvised and overcame. Move along, go about your business now.

You need to be logged in to post a comment. Login via the left margin or if you're new, register here.

The Consensus Project Website


(free to republish)

© Copyright 2024 John Cook
Home | Translations | About Us | Privacy | Contact Us