1. Wondering Aloud at 01:17 AM on 23 April 2008
   Shaviv claims that neutrons are the wrong thing to measure and Sloan basically answered "that's all we have", Did I read that right?
   
   If Shaviv is right and there is little magnetic field effect above 15 GeV, than neutrons are the wrong thing to measure. I don't know what they claimed about their study but it wouldn't be a representative test for the affect of cosmic rays on cloud formation.
   [ Response: Sloan's main point is that his analysis comes to the same result whether he uses neutron count or cosmic ray induced ionisation (CRII) rates. Equation 3 equates the cosmic ray component of cloud cover change to a constant K (kappa) multiplied by dN/N - the fractional change in neutron count N. So long as the CRII modulation has the same shape as neutron modulation, equation 3 is still valid.
   
   When Sloan and Wolfendale compared the neutron modulation to Usoskin's CRII data, they were surprised to find that not only was the shape compatible, the absolute values of the fractional change were also. This was why I asked Terry Sloan if he could email me a plot of the comparison of the fractional change in neutron count to CRII - so I could see it for myself.
   
   The bottom line is that both CRII and neutron counts increase at high latitudes so you would expect cloud cover amplitude to also increase. That cloud cover amplitude shows little dependence on latitude indicates cosmic rays are not a major factor in the modulation of cloud cover - if correlation between the 11 year signal in cloud cover and the solar cycle is real, most of it is due to other aspects of solar activity. ]
2. shawnhet at 06:44 AM on 23 April 2008
   I really don't see that Sloan's response addresses Shaviv's point at all. Presuming Shaviv is right (from his blog)
   
   "Another option is to calculate the actual latitudinal dependence of the atmospheric ionization variations. This was done by Usoskin et al. (2004), who took the top-of-the-atmosphere variations in the CRF, and using a *code* to calculate the shower products, calculated the actual latitudinal ionization rate variations.
   
   They found that the relative change in the LCC is the same as the relative change in the ion density (which itself is proportional to the square root of the ionization rate). *Both vary by several percent from equator to pole over the solar cycle.* This can be seen in fig. 2. In other words, the latitudinal dependence of the cloud cover variations is totally consistent with the CRF/cloud cover mechanism. *For comparison, the solar cycle variation in the neutron monitor data is almost 20% at the poles, and 5% at the equator.* "
   
   (asterisks mine)
   
   Since Sloan doesn't dispute this that I can see, he is using the wrong numbers (his proportion is off) to evaluate Shaviv's claims.
   
   Cheers, :)
   [ Response: Figure 3 compares the fractional change between neutron counts (solid dots and solid line) & ionisation (open squares, joined by the dashed line). The ionisation modulation is computed for solar cycle 22 for the total ionization from figure 5 of the Usoskin paper. It's the fractional modulation for cycle 22 multiplied by 2 since their fig 6 is a plot of n (which is proportional to the square root of q hence dq/q=2 dn/n, see equ 5 of their paper).
   
   This Usoskin paper where Sloan gets his ionisation data is "Cosmic ray-induced ionization in the atmosphere: spatial and temporal changes" published in Journal of Atmospheric and Solar-Terretrial Physics vol 66 (2004) page 1796, presumably the work Usoskin 2004 is based on. I haven't managed to get hold of this paper yet but haven't seen any criticism of the handling of this data from Shaviv or anywhere.
   
   The quibbling over neutrons versus ionisation is missing the major point that cosmic ray theory predicts latitude dependence in cloud cover modulation. The crucial result from Sloan and Wolfendale's paper is that cloud cover modulation shows little latitude dependence. ]
3. Quietman at 07:45 AM on 23 April 2008
   I have to guess that this would come under "other" in Cloud_formation_processes.
   I have seen articles and blogs all claiming that the cosmic ray hypothesis is false. Is there actually any agreement?
4. Robert S at 16:58 PM on 23 April 2008
   Here are a few studies that are in favor of the GCR-climate/cloud formation link (sorry the order is kind of a mess and you have probably addressed a few of them):
   
   http://arxiv.org/pdf/physics/0409123
   
   http://journals.royalsociety.org/content/77543w3q4mq86417/
   
   http://www.dsri.dk/~hsv/prlresup2.pdf
   
   http://www.spacecenter.dk/publications/scientific-report-series/...
   
   http://www.blackwell-synergy.com/doi/abs/10.1111/j.1468-4004...
   
   http://journals.royalsociety.org/content/3163g817166673g7/fulltext.pdf
   
   https://utd.edu/nsm/physics/pdf/Atmos_060302.pdf
   
   http://www.atmos-chem-phys.org/4/2273/2004/acp-4-2273-2004.html
   
   http://www.agu.org/pubs/crossref/2002/2001JA000248.shtml
   
   http://www.gsajournals.org/archive/1052-5173/13/7/pdf/i1052-5173-13-7-4.pdf
   
   http://www.utdallas.edu/physics/pdf/tin_atmtrans.pdf
   
   These two studies came out at roughly the same time as the S&W paper:
   http://aps.arxiv.org/ftp/arxiv/papers/0803/0803.2765.pdf
   http://aps.arxiv.org/ftp/arxiv/papers/0803/0803.2766.pdf
5. chris at 23:25 PM on 23 April 2008
   There are two questions really:
   
   (i) do cosmic rays influence clouds formation?
   (ii) is this effect, if real, significant with respect to the Earth's surface temperature and climates?
   
   Robert posted a list of papers that address (i) especially. However I'd say that there are some problems:
   
   The Shaviv Veizer analysis of a possible link between the CRF and Earth's temperature on the 100's of millions of years timescale (the GSA Today 2003 paper in Robert's list) rested on a paleo-CRF flux analysis and paleotemperature analysis. However Veizer has reinterpreted his temperature data (see abstract below ***) and the putative correlation that Shaviv described in the GSA article is now lost. In other words there is no evidence for a very-long-scale link between the CRF and Earth's temperature (and in fact the Earth's temperature right throughout the last around 500 million years correlates with the atmospheric CO2 level as Veizer indicates in his recent reinterpretation of paleotemperature data)
   
   Many of the papers in Robert's list are essentially neutral with respect to evidence for a CRF-cloud link. Thus Svensmarks Proc. Royal. Soc paper on micro-aerosol formation as a result of ionizing radiation in the lab is not an indication that CRF nucleates clouds in the real world...and so on...
   
   However there does seem to be some evidence (some of papers in Robert's list again) that the CRF can influence cloud formation (e.g. the Forbush events correlating transient changes in CRF with transient local changes in clouds).
   
   How about (ii) (significance with respect to the Earth's surface temperature)?
   
   This seems rather difficult to justify for several reasons. First, there isn't a correlation of CRF with the Earth's surface temperature on any timescale as far as I'm aware (see point about the Shaviv/Veizer analysis above). Second there has been no persistent trend in the CRF during the very marked warming of the last 30-odd years. Svensmark in his website report (http://www.spacecenter.dk/publications/scientific-report-series/Scient_No._3.pdf/view) explicitly demonstrates that the solar contribution (CRF et al!)during the very marked warming of the last 30-odd years is zero (a slight cooling contribution since 1958 if anything). That's pretty uncontroversial.
   
   The third general problem relates to the demonstration of the transient changes in local cloud cover in the so-called Forbush events that a couple of papers in Robert's list refer to. How can these processes give rise to a persistent long lived effects? This could presumably only occur with long lived persistent changes in the CRF. Short term local events (like those described as Forbush events if I understand these correctly) won't have much influence. That's partly the thrust of the Sloan/Wolfendale analysis. When the latter authors assess the relationship between Forbush events and cloud data averaged over weeks/months there isn't any correlation at all. Now one could argue that one should analyse only the correlations betwen transient changes in CRF and any transient local changes in clouds. But then the CRF/climate relationship becomes essentially unsupportable, because if there isn't any influence on a weekly/monthly averaged basis, how can there be significant persistent effects on temperature/climate?
   
   One explanation for the lack of correlation that might be significant for CRF-cloud-temperature/climate relationship is that the local nucleation of clouds mediated by CRF might relate to super-saturated regions of the atmosphere, which would anyway condense out into clouds by other (non-CRF) atmospheric nucleating species that are known to exist...
   
   Overall I would say that the evidence supports the possibility of some CRF influence on cloud formation but this remains to be substantiated, and that the evidence supports the conclusion that these effects are sufficiently small that they haven't provided any evidence in the real world for a CRF-cloud-temperature/climate link.
   
   
   Here's the recent Veizer reinterpretation of paleotemperature data:
   
   ***Came RE, Eiler JM, Veizer J et al (2007) Coupling of surface temperatures and atmospheric CO2 concentrations during the Palaeozoic era Nature 449, 198-201.
   
   Abstract: Atmospheric carbon dioxide concentrations seem to have been several times modern levels during much of the Palaeozoic era (543-248 million years ago), but decreased during the Carboniferous period to concentrations similar to that of today(1-3). Given that carbon dioxide is a greenhouse gas, it has been proposed that surface temperatures were significantly higher during the earlier portions of the Palaeozoic era(1). A reconstruction of tropical sea surface temperatures based on the delta O-18 of carbonate fossils indicates, however, that the magnitude of temperature variability throughout this period was small(4), suggesting that global climate may be independent of variations in atmospheric carbon dioxide concentration. Here we present estimates of sea surface temperatures that were obtained from fossil brachiopod and mollusc shells using the 'carbonate clumped isotope' method(5)-an approach that, unlike the delta O-18 method, does not require independent estimates of the isotopic composition of the Palaeozoic ocean. Our results indicate that tropical sea surface temperatures were significantly higher than today during the Early Silurian period (443-423 Myr ago), when carbon dioxide concentrations are thought to have been relatively high, and were broadly similar to today during the Late Carboniferous period (314-300 Myr ago), when carbon dioxide concentrations are thought to have been similar to the present-day value. Our results are consistent with the proposal that increased atmospheric carbon dioxide concentrations drive or amplify increased global temperatures(1,6).
   
6. Wondering Aloud at 03:05 AM on 24 April 2008
   That part I understood I think (re: response in 1). What Shaviv is saying is that for cosmic rays above 15 GeV there should not be increased incidence near the poles. (He doesn't say so but if the magnetic field is not significantly displacing them incidence near the poles should be reduced compared to lower latitudes) So cloud formation near the poles should not be increased if Cosmic Rays are a significant cause. So the entire premise of looking at cloud formation by latitude has nothing to do with the issue.
   
   The idea that cosmic rays cause cloud formation is not coming from this debate but rather from seperate direct experiments that clearly show it can do just that.
   
   
7. Robert S at 16:39 PM on 24 April 2008
   Chris (Colose?),
   I see your main issue is with Veizer's supposed "reinterpretation" of the data.
   
   Veizer's 2000 paper "Evidence for decoupling of atmospheric CO2 and global climate during the Phanerozoic eon."
   
   Veizer and Shaviv's 2004 paper "Celestial driver of Phanerozoic climate?"
   
   The 2007 Veizer paper you reference "Coupling of surface temperatures and atmospheric CO2 concentrations during the Palaeozoic era."
   
   Notice the difference? The paper your reference is describing coupling during the Palaeozoic era. The Palaeozoic era occurred during the Phanerozoic eon, which is what the 2000 and 2004 papers are about. I don't see any major reinterpretations, rather I see that Veizer found *some* evidence for coupling of CO2 and climate during roughly 1/3 of the Phanerozoic eon.
   
   An approximate one-third is still a thorn in the sides of the two earlier Veizer papers, but it certainly isn't enough to come to your conclusion of "...and the putative correlation that Shaviv described in the GSA article is now lost. In other words there is no evidence for a very-long-scale link between the CRF and Earth's temperature."
8. shawnhet at 01:42 AM on 25 April 2008
   "[ Response: Figure 3 compares the fractional change between neutron counts (solid dots and solid line) & ionisation (open squares, joined by the dashed line). The ionisation modulation is computed for solar cycle 22 for the total ionization from figure 5 of the Usoskin paper. It's the fractional modulation for cycle 22 multiplied by 2 since their fig 6 is a plot of n (which is proportional to the square root of q hence dq/q=2 dn/n, see equ 5 of their paper).
   
   This Usoskin paper where Sloan gets his ionisation data is "Cosmic ray-induced ionization in the atmosphere: spatial and temporal changes" published in Journal of Atmospheric and Solar-Terretrial Physics vol 66 (2004) page 1796, presumably the work Usoskin 2004 is based on. I haven't managed to get hold of this paper yet but haven't seen any criticism of the handling of this data from Shaviv or anywhere.
   
   The quibbling over neutrons versus ionisation is missing the major point that cosmic ray theory predicts latitude dependence in cloud cover modulation. The crucial result from Sloan and Wolfendale's paper is that cloud cover modulation shows little latitude dependence. ]"
   
   Well, first off, whether there is latitude dependence ,is dependent on what latitude dependence looks like(Shaviv thinks it will look different than Sloan). Secondly, if you look at your third figure there you will see that the slope of the muon line is much less than the neutron monitor line(consistent with what Shaviv said). Since, IIRC, muons are the particles that are hypothetically responsible for lower atmosphere atnospheric ionisation, we should be basing our conclusions on that line not the other two.
   
   There appears to be some confusion over what Usoskin is actually saying. Both Shaviv and Sloan think Usoskin supports their conclusions. Possibly, one or the other, has misinterpreted the Usoskin results.
   
   I found the paper in question, but I must admit it is outside my ability to actually move from there to the calculations in question.
   
   http://cc.oulu.fi/~usoskin/personal/JASTP_published.pdf
   
   "The ionization of the atmosphere at low and moderate
   altitudes is fulfilled not by the primary CR particles but by secondaries of a nucleonic–electromagnetic cascade initiated by primary energetic cosmic rays in the Earth’satmosphere. Accordingly, in order to study the cosmic ray-induced ionization, one needs to take into account the development of such a cascade. Here we employed the CORSIKA Monte Carlo package (Heck et al., 1998) which is specially designed to simulate cascade and includes recent and reliable description of various
   physical processes and cross-sections. Cosmic rays are
   assumed to consist of protons and a-particles (
   6% in
   particle number). (When denoting CR energy we mean
   energy per nucleon, throughout the paper.) In particular, CORSIKA can calculate energy losses deposited by the developing cascade for ionization of the ambient air at every step."
   
   Now, I don't know what that means exactly ;) , but it sure sounds more complex than a simple proportion.
   
   Also, just curious, since you agree that there is solar effect on climate, how do you think that this effect happens. if not at least partly by way of cosmic rays?
   
   Cheers, :)
9. chris at 02:48 AM on 25 April 2008
   Hi Robert,
   
   I wouldn’t say it’s my main issue (the absence of trends in CRF during the period of large contemporary warming is a serious issue too)! However it is surely a serious flaw with respect to the notion of a dominant influence of the CRF on the Earth’s temperature during the last 500 million years.
   
   Yes, Veizer’s reanalysis applies specifically to the Palaeozoic. However it basically negates the putative relationship between Shaviv’s putative CRF cycle and the Earth’s temperature over one full cycle (i.e. where the putative CRF indicates a cold period centred around 450 MYA, Veizer himself now indicates that the earth was very warm; compare Figure 2 of Shaviv and Veizer (2003), with Figure 2 of Carne, Eiler, Veizer et al (2007). Veizer’s reanalysis indicates that he himself considers his early analyses based on fossil carbonate del-18O were problematic. Presumably other periods of Veizer’s temperature reconstruction may similarly be problematic as indicated by other independent studies (see following); e.g. as described in Royer et al (2004).
   
   I really only highlighted Veizer’s reassessment of the Palaeozoic paleotemperature since it was Veizer’s own temperature reconstruction that Shaviv fitted his putative CRF cycles to. However there are many other independent problems with the putative correlation. For example if one examines the entire paleotemperature record and paleoCO2 record (as done for example in Royer’s recent compilations at least with respect to identification of well-defined evidence for significant glacial episodes), the cold periods expected in the Mesozoic (centred around 165 MYA according to Shaviv’s putative CRF reconstruction) aren’t represented in the record (see Royer 2006 reference below).
   
   If a major chunk of the Shaviv-Veizer apparent “correlation” is “dis-correlated” (!) by Veizer himself then there’s clearly a problem as is indicated by other independent proxy-temperature data.
   
   There’s a more general problem I think which relates to the seductive ease of fitting very grand “cycles” to extremely sparse data sets (paleotemperature/paleoCRF, although the latter is implicitly “cyclic”). With more abundant data (e.g. the Royer 2006 compliation of coincident paleoCO2 and paleo”temperature” data) the historical records look rather more realistic – they don’t seem to be cyclic at all. There are warm and cold periods, and the warm periods are interrupted by spikes of mild, medium or strong glaciations and so on. Now it is possible that the slow passage of the Earth through the long arms of our galaxy has some influence on climate. But it doesn’t seem very convincing in the record, and in fact that data seems to support a dominant temperature-CO2 relationship (as Veizer’s own recent work supports).
   
   R.E. Carne, J.M. Eiler, J. Veizer et al (2007) Coupling of surface temperatures and atmospheric CO2 concentrations during the Palaeozoic era; Nature 449, 198-202
   
   D. L. Royer et al (2004) CO2 as a primary driver of Phanerozoic climate; GSA Today March 2004 pp 4-10
   
   D. L. Royer (2006) CO2-forced climate thresholds during the Phanerozoic Geochim. Cosmochim Acta 70 5665-5675
   
   Shaviv, N.J. and Veizer, J. (2003) Celestial driver of Phanerozoic climate?; GSA Today July 2003 pp 4-9
   
   P.S. I’m a common or gardener “Chris” and not “Chris Colose”!
   
10. chris at 03:00 AM on 25 April 2008
    shawnhet:
    
    re your comment: "Also, just curious, since you agree that there is solar effect on climate, how do you think that this effect happens. if not at least partly by way of cosmic rays?"
    
    Surely the most likely solar contributions to the Earth's heat budget (and thus climate) is the solar irradiance and related parameters. It's very clear that this (or a related parameter that is reflected in the sunspot data for which there is a very good several hundred year old record) can cause very significant effects on the Earth's temperature as indicated by the cold periods around the Maunder Minimum and the possible contribution of rising solar outputs during the early decades of the 20th century.
    
    Now it may be the cause that some of these effects were due to solar influences on the cosmic ray flux (CRF). Unfortunately it's not easy to seperate these contributions out. However there isn't really any compelling evidence for a CRF influence on the Earth's surface temperature. When someone comes up with some evidence then I'm sure that we'll all believe it!
    
    What we can say is that there's no contribution from the CRF to the widespread and rather significant contemporary warming (since the early 70's). Even the most ardent supporters of the CRF idea consider that to be the case.
    
11. shawnhet at 04:33 AM on 25 April 2008
    Hi Chris,
    
    If a direct solar input (say TSI) causes the solar cycle trend in cloudiness, that would imply that increased cloudiness is a function of increased water vapor in the atmosphere correct? It is pretty hard to see how decreased water vapor can lead to increased cloudiness. The problem with this is that the direction is the wrong way around, isn't it? We would expect there to be more water vapor in the atmosphere, when it was warmer, hence, it should be easier to become cloudy when the sun is hotter.
    
    IAC, are you aware of any studies that compare TSI to cloudiness in the lower atmosphere? I would be interested in these if they exist.
    
    "Now it may be the cause that some of these effects were due to solar influences on the cosmic ray flux (CRF). Unfortunately it's not easy to seperate these contributions out. However there isn't really any compelling evidence for a CRF influence on the Earth's surface temperature. When someone comes up with some evidence then I'm sure that we'll all believe it!"
    
    Well, I guess that this is a matter of opinion, but I think that the CRF hypothesis is better than any alternative. It does have some actual empirical support for the formation of cloud condensation nuclei.
    
    It is entirely possible that the CRF effect is not very large or (more likely) been swamped by some other effect in recent years. The latter would probably be the *most* consistent with *all* the evidence.
    
    Cheers, :)
12. Robert S at 16:46 PM on 25 April 2008
    Chris,
    
    The Came et al 2007 paper uses a (relatively) new temperature reconstruction method and Veizer notes here that there is still much uncertainty: http://blogs.nature.com/nature/journalclub/2007/10/francis_albarede.html
    
    It is yet to be determined if this method is more accurate. Celestial drivers better explain the GEOCARB Phanerozoic reconstructions, while CO2 fits (somewhat) well with the Came 2007 reconstruction.
    
    Regarding Royer 2004, Shaviv and Veizer gave this comment: http://www.gsajournals.org/pdf/online_forum/i1052-5173-14-3-e4.pdf
    
    A quote:
    "Note that Royer et al. do not dispute the existence of the CRF/temperature correlation of Shaviv and Veizer (2003), only its role relative to that of CO2."
    
    Despite the Came 2007, the paper still applies... for the most part.
13. chris at 05:29 AM on 28 April 2008
    Shawnet,
    I was adressing your comment:
    
    “re your comment: "Also, just curious, since you agree that there is solar effect on climate, how do you think that this effect happens. if not at least partly by way of cosmic rays?"
    
    My point is that there is zero evidence for a cosmic ray flux (CRF) effect on climate (can you give me some evidence in support of this notion?), and that a solar effect on climate is very easily understood in relation to solar irradiance (or other direct changes in the intensity of solar electromagnetic radiation impacting on the Earth.
    
    So in response to your question:
    
    “If a direct solar input (say TSI) causes the solar cycle trend in cloudiness, that would imply that increased cloudiness is a function of increased water vapor in the atmosphere correct?”
    
    I would say, first, that it hasn’t got anything necessarily to do with “cloudiness”. Why bring clouds into it at all? Surely the dominant solar influence is irradiance (thermal energy!). If the sun burns hotter (higher irradiance) then the Earth will receive a higher flux of UV, IR, light intensity and will be warmer. Why bring clouds into it at all?
    
    And second, why should increased cloudiness necessarily be “a function of increased water vapour in the atmosphere”? After all, we know that when the atmosphere warms, the water vapour concentration increases (simple physics of water partitioning between an aqueous pool; i.e. the oceans, and the atmosphere). However a warmer atmosphere has a higher saturation point for water vapour. So why should there be more clouds??
    
    And when you say:
    
    
    “Well, I guess that this is a matter of opinion, but I think that the CRF hypothesis is better than any alternative. It does have some actual empirical support for the formation of cloud condensation nuclei.”
    
    Which is in response to my statement that it’s not easy to separate CRF effects from total solar irradiance effects and so on, since these parameters are generally correlated, and certainty so within the 11 year solar cycle….
    
    …I would ask, what is the evidence that the CRF “is better than any alternative”? After all, you are presuming that a doubly uncharacterised effect (the putative CRF effect on clouds that might have an effect on the Earth’s surface temperature) is more significant than a direct total solar irradiance effect that we know has an effect on the Earth’s surface temperature (se the Maunder Minimum and its effects, or the early 20th century change on solar output and it’s likely contribution to early 20th century warming.
    
    It’s really a question of the evidence. I think we can agree that the CRF is not very large. There's certainly not evidence that indicates otherwise.
    
14. shawnhet at 05:56 AM on 28 April 2008
    Hi Chris,
    
    Sorry, reading more carefully, I see that you were likely ascribing temperature variations to TSI, not cloudiness variations(as stated in the opening paragraph). Given that I assume that there is a correlation btw solar proxies of various sorts, my point is that CRF is the most logical cause of that relationship(TSI on its own doesn't change the prevalence of cloud building blocks except by putting a bit more moisture in the air). If you don't accept that there is a relationship btw cloudiness and solar proxies, we are simply talking past each other. My apologies for the confusion.
    
    In re, the point about how easy it is to separate CRF from TSI and other factors, it is pretty difficult to do so for the *last few decades* I agree. However, I do think that one can correlate cloudiness with CRF, which does a better job of explaining past behavior, (eg the various solar Minimums), than TSI on its own. TSI on its own doesn't change by enough Joules to explain the MM!
    
    There may be something peculiar that happened btw 1970-2000 that overwhelmed/hid the CRF effect, but it remains the best explanation for the *magnitude* of the climate changes during the past. Ultimately, though, the science is still not conclusive.
    
    Cheers, :)
15. chris at 06:21 AM on 28 April 2008
    Hi Shawnet,
    
    fair enough about your comments. But I have a problem with this:
    
    "There may be something peculiar that happened btw 1970-2000 that overwhelmed/hid the CRF effect, but it remains the best explanation for the *magnitude* of the climate changes during the past. Ultimately, though, the science is still not conclusive. "
    
    Of course there is "something peculiar" that happened betwen 1970-2000. It's the massive enhancement of the Earth's greenhouse effect, especially since the massive post-war increase in greenhouse gas emissions that kicked in during the early 1960's.
    
    But how can you say that it (the CRF effect) is "the best explanation for the *magnitude* of the climate changes during the past". What's the evidence upon which such an assertion is based??? That's the point... The evidence.
    
    If one examines the complete record of contemporaneous proxy-temperature data and proxy-CO2 data, the entire Phanerozoic period (last 500 million years or so) highlights a rather good relationship between the Earth's surface temperature (especially in relation to evidence for mild, medium and strong glaciations) and atmopsheric CO2 [e.g. see Royer DL (2006)]. There is no particularly well-validated relationship between the supposed CRF and the Earth's surface temperature during this period at all.
    
    So why bother to assert a relationship with the supposed variations in CRF its supposed effect on the Earth's surface temperature (for which there isn't any particular evidence)...
    
    ...in short...where's the evidence???
    
    
    
    Royer, DL (2006) CO 2-forced climate thresholds during the Phanerozoic. Geochim. Cosmochim. Acta 70, 5665–5675
    
    
16. chris at 06:33 AM on 28 April 2008
    Robert,
    
    There seems to be something akin to “conspiracy theory” in the approach of the Cosmic Ray Flux (CRF) hypothesisers!
    
    You state that Shaviv and Veizer say that:
    “A quote:
    "Note that Royer et al. do not dispute the existence of the CRF/temperature correlation of Shaviv and Veizer (2003), only its role relative to that of CO2."
    
    That seems a specious comment to me, since it’s rather clear on re-reading Royer et al (2004), that their absence of “dispute” of the “existence of the CRF/temperature correlation” isn’t some sort of implicit acceptance of the supposed correlation at all! They don’t particularly comment on the CRF/temperature “correlation” since (i) the CRF/temperature “correlation” isn’t really worthy of comment since there’s little evidence that it exists, and (ii) Royer et al direct the reader to an article that rather carefully highlights the gross deficiencies in the Shaviv/Veizer hyopothesis [i.e. Rahmstorf et al (2004)].
    
    Now there may be a cyclic variation in the CRF relating to the slow passage of the Earth through the spiral arms of our galaxy, but Shaviv and Veizer don’t present very compelling evidence for such a phenomenon. This putative cyclic variation in the CRF might have a period equivalent to that suggested by Shaviv and Veizer but the evidence for such a period is not compelling. For example, Shaviv and Veizer, in attempting to assign a period to the putative CRF flux inferred from meteorite “clusters”, use as one of their “parameters” “that which best fits the ice age epochs” (see Figure 10 of Shaviv and Veizer, 2003, GSA Today). In other words they use the climate history as a parameter to “fit” their CRF cycles, and then proceed to conclude that the CRF “fits” the Earth’s climate history. That’s just poor science.
    
    In the meantime Veizer has reinterpreted his paleotemperature data (for the Paleozoic) and concluded that this fits rather better to the paleoCO2 data (and gives a large “dis-correlation” between the supposed CRF cycle, for which little evidence actually exists! and the paleotemperature data) (Came et al, 2007) , and a wealth of independent proxy paleoCO2 and paleotemperature data provides a non-cyclic variation in the Earth’s surface temperature throughout the Phanerozoic in which the dominant correlate is the atmospheric CO2 concentration. This is summarized in Royer’s very recent compilation (Royer, 2006).
    
    What’s concerning about this whole area is the manner in which its advocates engage in trying to bypass the scientific arena to play directly to the “peanut gallery”! Henrik Svensmark is particularly prominent in this dodgy practice. His rather limited and poorly substantiated publications on the CRF/clouds/climate relationship are all very well (maybe the relationship exists…maybe it doesn’t…the evidence is rather unsubstantial), but he then uses these to play fast and loose with a dodgy and rather unscientific book, and a ludicrous website report that is a complete travesty of the scientific method (whatever that might be!....however we can recognise, I hope, that the Svensmark and Friis Christensen (2007) web site “report” that you urled is complete pants).
    
    If it’s science it must be about evidence…..
    
    
    
    Came, R.E., J.M. Eiler, J. Veizer et al (2007) Coupling of surface temperatures and atmospheric CO2 concentrations during the Palaeozoic era; Nature 449, 198-202
    
    Rahmstorf, S 2004, Cosmic rays, carbon dioxide, and climate: Eos (Transactions, American Geophysical Union), 85, p. 3841.
    
    Royer, D. L. et al. (2004) CO2 as a primary driver of Phanerozoic climate
    GSA Today March 2004 pp 4-10.
    
    Shaviv, N.J. and Veizer, J. (2003) Celestial driver of Phanerozoic climate?; GSA Today July 2003 pp 4-9
    
    Svensmark and Friis-Christensen (2007) (website “report”
    http://www.spacecenter.dk/publications/scientific-report-series/...
    
17. Robert S at 17:16 PM on 28 April 2008
    Chris said:
    "That seems a specious comment to me, since it’s rather clear on re-reading Royer et al (2004), that their absence of “dispute” of the “existence of the CRF/temperature correlation” isn’t some sort of implicit acceptance of the supposed correlation at all! They don’t particularly comment on the CRF/temperature “correlation” since (i) the CRF/temperature “correlation” isn’t really worthy of comment since there’s little evidence that it exists, and (ii) Royer et al direct the reader to an article that rather carefully highlights the gross deficiencies in the Shaviv/Veizer hyopothesis [i.e. Rahmstorf et al (2004)]."
    
    It was not an "implicit acceptance," but it wasn't an implicit refutation of the CFR-climate link either, rather it was an attempt to explain CO2s relative role
    
    Shaviv gives this summary of his and Veizer's response to Royer:
    "[Royer et al.] argued that the 18O/16O based temperature reconstructions (of Veizer et al. 2000) has an unaccounted systematic error, due to ocean pH, and hence the atmospheric pCO2 level. Shaviv (2005) considered this effect and showed that instead of an upper limit to the effect of CO2 doubling, of 1°C, Earth's sensitivity increases to 1-1.5°C, but the basic conclusion that CRF appears to be the dominant climate driver remains valid (as later independently confirmed by Wallman 2004)"
    and
    "Incidentally, there was a more serious attack on our work, by Royer et al., but it was not about the validity of the CRF/climate correlation. Instead, it discussed the fact that CO2 could offset the geochemical temperature reconstruction. If you take that correction into account and redo the analysis (which Royer et al. didn't) you find that CO2 had a larger role, but still secondary to the cosmic ray flux over geological time scales."
    
    Regarding Rahmstorf 2004, Shaviv summarizes his response to them here and why he thinks their response to his rebuttal fails:
    "every single point they raise is invalid... in their reply to it ["it" referring to Shaviv's rebuttal], they don't address any of the points and simply discuss the statistical meaning of the cosmic ray flux / temperature correlation. In our rebuttal to that, you'll find why their statistical analysis grossly fails, because they unknowingly used Bartlett's formula in a limit where its basic assumption is invalid. In fact, if you redo their statistical analysis without this gross mistake, you realize that the statistical significance of the CRF temperature correlation is at least at the 99.7% level (and this is without the sedimentation or astronomical records)...It is in fact the most significant correlation between any climate variable and a radiative forcing proxy on a time scale longer than a few million years. Moreover, the CRF data and the 18O data are backed with additional, independent data sets, making the link redundant and robust. It implies, again, that the CRF was the dominant climate driver on the multimillion year time scale."
    
    Here is Rahmstorf 2004 paper:
    http://www.agu.org/pubs/crossref/2004/2004EO040002.shtml
    
    And here is paper that was in response to Rahmstorf 2004: http://www.phys.huji.ac.il/~shaviv/ClimateDebate/RahmstorfDebate.pdf
    
    And here is the Shaviv and Veizer paper that was in response to Rahmstorf's reply to S&V's response to Rahmstorf 2004 (that was a mess): http://www.phys.huji.ac.il/%7Eshaviv/ClimateDebate/RahmReplyReply.pdf
    
    On the other hand, Wallman essentially agreed with Shaviv and Veizer: http://www.agu.org/pubs/crossref/2004.../2003GC000683.shtml
    
    You then say:
    "For example, Shaviv and Veizer, in attempting to assign a period to the putative CRF flux inferred from meteorite “clusters”, use as one of their “parameters” “that which best fits the ice age epochs” (see Figure 10 of Shaviv and Veizer, 2003, GSA Today). In other words they use the climate history as a parameter to “fit” their CRF cycles, and then proceed to conclude that the CRF “fits” the Earth’s climate history. That’s just poor science."
    
    That is quite an accusation. So, if you don't mind me asking, do you have evidence to support this?
    
    You say
    "maybe the relationship exists…maybe it doesn’t…the evidence is rather unsubstantial."
    
    Unsurprisingly, this is where advocates of the CRF/climate theory would disagree with you, and you can find out why by the list of papers I provided above (and there are several I missed).
    
    And here is Shaviv's brief summary on the evidence for the link:
    
    "Svensmark (1998) finds that there is a clear correlation between cosmic rays and cloud cover. Since the time he first discovered it, the correlation continued as it should (Svensmark, 2007). Here is all the other evidence which demonstrates that the observed solar/cloud cover correlation is based upon a real physical link.
    
    1) Empirical Solar / CRF / Cloud Cover correlation: In principle, correlations between CRF variations and climate does not necessarily prove causality. However, the correlations include telltale signatures of the CRF-climate link, thus pointing to a causal link. In particular, the cloud cover variations exhibit the same 22-year asymmetry that the CRF has, but no other solar activity proxy (Fichtner et al., 2006 and refs. therein). Second, the cloud cover variations have the same latitudinal dependence as the CRF variations (Usoskin et al. 2004). Third, daily variations in the CRF, and which are mostly independent of the large scale activity in the sun appear to correlated with cloud variations as well (Harrison and Stephenson, 2006).
    
    2) CRF variations unrelated to solar activity: In addition to solar induced modulations, the CRF also has solar-independent sources of variability. In particular, Shaviv (2002, 2003a) has shown that long term CRF variations arising from passages through the galactic spiral arms correlate with the almost periodic appearance of ice-age epochs on Earth. On longer time scales, the star formation rate in the Milky Way appears to correlate with glacial activity on Earth (Shaviv, 2003a), while on shorter time scale, there is some correlation between Earth magnetic field variations (which too modulate the CRF) and climate variability (Christl et al. 2004).
    
    3) Experimental Results: Different experimental results (Harrison and Aplin, 2001, Eichkorn et al., 2003, Svensmark et al. 2007) demonstrate that the increase of atmospheric charge increases the formation of small condensation nuclei, thus indicating that atmospheric charge can play an important role (and bottleneck) in the formation of new cloud condensation nuclei.
    
    4) Additional Evidence: Two additional results reveal consistency with the link. Yu (2002), carried out a theoretical analysis and demonstrated that the largest effect is expected on the low altitude clouds (as is observed). Shaviv (2005) empirically derived Earth's climate sensitivity through comparison between the radiative forcing and the actual temperature variations. It was found that if the CRF/cloud cover forcing is included, the half dozen different time scales which otherwise give inconsistent climate sensitivities, suddenly all align with the same relatively low climate sensitivity, of 0.35±0.09°K/(W/m2)."
    
    You can find more by Shaviv here: http://www.sciencebits.com/SiteContent
    
    I apologize for all the cut and paste, but if I am going to represent a theory, it might be best if I let those who know something about it, explain it.
    
    Of course cosmic rays aren't meant to explain ALL of the warming, rather they are meant to represent a large piece of the puzzle, and with other things (including human activities and CO2), attempt to explain this warming.
    
    Regards,
    
    Robert
    
18. chris at 20:32 PM on 28 April 2008
    Robert,
    
    In response to my comment:
    
    "For example, Shaviv and Veizer, in attempting to assign a period to the putative CRF flux inferred from meteorite “clusters”, use as one of their “parameters” “that which best fits the ice age epochs” (see Figure 10 of Shaviv and Veizer, 2003, GSA Today [P.S. I meant Figure 2!]). In other words they use the climate history as a parameter to “fit” their CRF cycles, and then proceed to conclude that the CRF “fits” the Earth’s climate history. That’s just poor science."
    
    you say:
    Robert: "That is quite an accusation. So, if you don't mind me asking, do you have evidence to support this?"
    
    
    It's not an "accusation"...it's a statement of fact. Shaviv and Veizer say exactly that in their GSA 2003 article.
    
    For example, here's the relevant text from the legend of their Figure 2 (see page 6 of Shaviv and Veizer, GSA Today, July 2003)
    
    "The blue line depicts the
    nominal CRF, while the yellow shading delineates
    the allowed error range. The two dashed curves are
    additional CRF reconstructions that fit within the
    acceptable range (together with the blue line, these
    three curves denote the three CRF reconstructions
    used in the model simulations). The red curve
    describes the nominal CRF reconstruction after its
    period was fine tuned to best fit the low-latitude
    temperature anomaly (i.e., it is the “blue”
    reconstruction, after the exact CRF periodicity was
    fine tuned, within the CRF reconstruction error)."
    
    In other words a "CRF periodicity" was assummed and then fitted to a set of metorite cluster data (which is decidely non-sinusoidally-periodic as it happens). This independent sinusoidal variation which doesn't fit the geological temperature record very well in the early part of the Phanerozoic, is readjusted ("fine tuned") "to best fit the low altitude temperature anomaly". It is this "fine tuned" putative cyclic CRF flux (the red curve in Shaviv and Veizer) that is then compared with the paleotemperature record.
    
    That seems pretty clear. It's rather clear that Shaviv and Veizer have pre-assummed a sinusoidal CRF flux (for which there isn't compelling evidence), and then assummed that this should match the paleotemperature record (no reason why it should necessarily), and then adjusted the period of their putative sinusoidal flux so that it does fit the paleotemperature record, and then argue that there is a correlation between a putative CRF flux ("fine tuned" to match the paleotemperature record) and the paleotemperature record.
    
    One wouldn't get away with that in the research field in which I work....however I expect that it's fair to say that Shaviv and Veizer prepared a rather provocative piece for publishing in the GSA house journal and gave it a title with a question mark; i.e. " Celestial driver of Phanerozoic climate?"
    
    The answer to the question would seem to be so far: probably not. The evidence is poor and the paleotemperature data, which actually doesn't have a nice sinusoidal variation at all) fits rather better to the paleoCO2 data as detailed in the recent very comprehensive compilation of Royer (2006; seee my posts above for the citation).
    
    As for your statement:
    
    "Of course cosmic rays aren't meant to explain ALL of the warming, rather they are meant to represent a large piece of the puzzle, and with other things (including human activities and CO2), attempt to explain this warming."
    
    Which "warming" are you referring to? Cosmic rays don't "explain" any of the recent warming (last 30-odd years) at all. Svensmark himself indicates that to be the case. Cosmic rays don't explain any of the warming of the last 150 years, nor any of the warming during the ice age cycles, nor any of the warming throughout the Earth's paleohistory and so on.
    
    Now cosmic rays might have some role. But there isn't any compelling evidence for such a role and we know rather categorically that they have played no significant role in the very marked warming of the last 30-odd years. Apart from everyone else, Svensmark says so....he could hardly say otherwise since the data is rather straightforward and compelling!
    
    
19. shawnhet at 02:10 AM on 29 April 2008
    "But how can you say that it (the CRF effect) is "the best explanation for the *magnitude* of the climate changes during the past". What's the evidence upon which such an assertion is based??? That's the point... The evidence.
    
    If one examines the complete record of contemporaneous proxy-temperature data and proxy-CO2 data, the entire Phanerozoic period (last 500 million years or so) highlights a rather good relationship between the Earth's surface temperature (especially in relation to evidence for mild, medium and strong glaciations) and atmopsheric CO2 [e.g. see Royer DL (2006)]. There is no particularly well-validated relationship between the supposed CRF and the Earth's surface temperature during this period at all."
    
    Hi Chris,
    
    There are many points to be made here, but I will constrain myself to just a few. The vast majority of papers on the cosmic ray-climate link do support the idea of such a link. The fact that something else might also be happening for the last 30% of the 20th Century doesn't mean that such a link is not substantial.
    
    for instance here,
    
    http://www.blackwell-synergy.com/doi/abs/10.1046/j.1468-4004.2000.00418.x?journalCode=aag
    
    rays and climate
    The influence of cosmic rays on terrestrial clouds and global warming
    E Pallé Bagó C J ButlerArmagh Observatory College Hill, Armagh, BT61 9DG, N. Ireland.
    Note: Following completion of this paper, the authors were informed by Drs N Marsh and H Svensmark, that they have obtained similar results to those presented here. Marsh and Svensmark’s paper will appear in a forthcoming issue of Space Science Review.
    
    Acknowledgements. The cloud D2 datasets were obtained from the NASA Langley Research Center EOSDIS Distributed Active Archive Center. The authors would like to thank K O'Brien, B Bromage and G McCormak for discussions and M Murphy for assistance with computing. Research at Armagh Observatory is grant-aided by the Department of Culture, Arts and Leisure for Northern Ireland.
    
    Abstract
    We analyse the new ISCCP (International Satellite Cloud Climatology Project) D2 cloud data to ascertain if a connection between cosmic-ray flux and cloud cover exists. Despite a previous finding that total-cloud factor and cosmic-ray fluxes were correlated, our results indicate that only the low-level cloud follows solar activity over the full period, 1983–94. Using several proxies for solar activity and the radiative forcing calculated by Ockert-Bell (1992) for the ISCCP cloud types, we estimate the possible impact that such a solar–terrestrial connection may have on climate. We conclude that, possibly excluding the most recent decades, much of the warming of the past century can be quantitatively accounted for by the direct and indirect effects of solar activity.
    
    No, single factor can explain all the climate variation we have seen over the last several years. CO2 changes don't explain the Solar minimum changes even if they do explain the last 1970-2000. A theory of climate should be able to explain both.
    
    Cheers, :)
20. chris at 07:28 AM on 29 April 2008
    shawnet,
    
    you state:
    
    "There are many points to be made here, but I will constrain myself to just a few. The vast majority of papers on the cosmic ray-climate link do support the idea of such a link. The fact that something else might also be happening for the last 30% of the 20th Century doesn't mean that such a link is not substantial."
    
    and then you cite an old paper by Palle Bego and Butler from a journal that isn't indexed in the scientific database.
    
    In fact the "vast majority of papers on the cosmic ray-climate link" DON'T support such a link even if they might support "the idea of such a link"! That's the problem. We're back in conspiracy theory mode...
    
    Have a read of Palle Bego and Butler [Astronomy and Geophysics 41, 4.18-4.22 (2000)]
    
    You'll see that it's a bit of a fishing expedition. The entire time series studied is 1983-1994. They examine a bit of this tiny temporal period and come up with an estimated relationship that they use to extrapolate back in time during periods where there is no cloud cover data (and previous to 1953 where there is no direct CRF data). In short they make a large number of unverified assumptions of which they state "It must be admitted that these assumptions are open to question." and "However, a number of assumptions have been made, which if proved wrong would invalidate these estimates."
    
    The fact is that whatever the relationship between the cosmic ray flux (CRF) and cloud formation, there has been no systematic trend in the CRF since the late 1950's and therefore there can have been no CRF contribution to the warming trend of the last 30-odd years. Even the most vocal advocate of the CRF hypothesis (Svensmark) has concluded this. His analysis indicates if anything a mild cooling contribution from solar contributions (CRF et al) since around 1958.
    
    In relation to your comment:
    
    "No, single factor can explain all the climate variation we have seen over the last several years. CO2 changes don't explain the Solar minimum changes even if they do explain the last 1970-2000. A theory of climate should be able to explain both."
    
    I'm not sure what you mean by this. Are you considering year-on-year variation? If so there is nothing mysterious, is there? Within a world undergoing a transition to a new equilibrium temperature under the influence of an enhanced greenhouse effect, one obviously doesn't expect a perfect regular temperature increment year by year! El Nino's, La Nina's, volcanic aerosols, the solar cycle and various stochastic elements of the climate system will all act to modulate the greenhouse enhanced temperature increase, sometimes countering this and sometimes adding to it. That's why one doesn't get too excited about temperature variations over periods of 1-3 or 4 years. 2007 was pretty much indistinguishable from 2005 and 1998 as the warmest years on record. 1998 was lifted by around 0.2 oC above the long term trend by the strongest El Nino of the 20th century. 2005 and 2007 got there without the "help" of a strong El Nino. Right now we're coming out of a La Nina event and we're smack at the bottom of the solar cycle. So we're not surprised if early 2008 has a coolish spell and we don't expect any records to be broken in 2008 and maybe not 2009. However it wouldn't be surprising if the next warmest year on record coincided with the next major El Nino event. Our current "theory of climate" accommodates all of that I would have thought!
    
    But there's nothing particulary odd or mysterious about any of this. Svensmark has explicitly ruled out CRF contributions to the Earth's energy budget since the late 1950's pretty much in line with all of the other solar observers. I don't quite see the point of trying to sneak CRF in by the back door by insinuations to various bits of poorly substantiated snippits of unsupported work.
    
    If there is some evidence for a CRF contribution to the Earth's heat budget then everyone will be more than happy to accept it I'm sure! After all if it's supported by the evidence then it will have to be taken seriously. At present the evidence is rather dismal, even to the extent that the greatest proponent of the hypothesis (Svensmark) demonstrated several months ago, presumably to his own satisfaction, that there's been no solar contribution (a slight cooling contribution since 1958 if anything) to the Earth's "temperature" since 1958...
21. shawnhet at 08:33 AM on 29 April 2008
    Chris, I think we are clearly talking past each other here. I concede that the last thirty years of the Twentieth Century are not well explained by the CRF. However, this does not imply that that the effect of CRF on climate is small. It just means that it can't explain the last 30 years on its own. Likewise, a 30 year period with no change in the frequency of earthquakes doesn't establish that earthquakes don't strongly impact climate either. There are also 30 year periods in recent history that are not well explained by an enhanced greenhouse.
    
    As to whether the majority of the studies support the connection, I will say that the majority that I have seen do so(as do the majority of the papers listed on this thread). We can obviously argue the merits of each individual paper, but I don't think this is a productive use of time.
22. Francois O at 11:03 AM on 29 April 2008
    I find it funny that when it comes to cosmic rays, anything but a perfect correlation is taken as a refutation. But for greenhouse gases:
    
    "El Nino's, La Nina's, volcanic aerosols, the solar cycle and various stochastic elements of the climate system will all act to modulate the greenhouse enhanced temperature increase, sometimes countering this and sometimes adding to it. "
    
    Like the previous post on this blog, where ENSO effects were removed, removing any warming for a 50 year period where greenhouse gases were increasing...
    
    I also find it strange that scientists who study the link between cosmic rays and climate are called "skeptics". Are they not just scientists? Are not those who claim there is no link between the Sun and climate themselves "skeptics"? This blog seems to be about skepticism, but it chooses to be one-sided in its skepticism. For some strange reason, all studies about cosmic rays have some fatal flaws, and there is never ever any error in any paper showing that GHG's are 100% responsible for all disasters past and future!
    
    Well, of course, there was the thermohaline circulation thing that was, ahem..., a bit premature... and the hurricane-warming connection that even Kerry Emmanuel has abandoned... were not those flawed "peer-reviewed" papers?
    
    The CR-climate link may be complex and subtle, as Usoskin latest paper has shown: there is a regional pattern to cosmic rays. The effect on clouds depends on location, not just latitude. Until the effect is fully understood, parameterized, and included in GCM's, it seems a bit futile to try to refute it with simplistic arguments, and then claim that similarly simplistic arguments should not be used to refute AGW.
    
    Just my two cents...
    
    
    
    
23. Francois O at 11:49 AM on 29 April 2008
    I find it funny that when it comes to cosmic rays, anything but a perfect correlation is taken as a refutation. But for greenhouse gases:
    
    "El Nino's, La Nina's, volcanic aerosols, the solar cycle and various stochastic elements of the climate system will all act to modulate the greenhouse enhanced temperature increase, sometimes countering this and sometimes adding to it. "
    
    Like the previous post on this blog, where ENSO effects were removed, removing any warming for a 50 year period where greenhouse gases were increasing...
    
    I also find it strange that scientists who study the link between cosmic rays and climate are called "skeptics". Are they not just scientists? Are not those who claim there is no link between the Sun and climate themselves "skeptics"? This blog seems to be about skepticism, but it chooses to be one-sided in its skepticism. For some strange reason, all studies about cosmic rays have some fatal flaws, and there is never ever any error in any paper showing that GHG's are 100% responsible for all disasters past and future!
    
    Well, of course, there was the thermohaline circulation thing that was, ahem..., a bit premature... and the hurricane-warming connection that even Kerry Emmanuel has abandoned... were not those flawed "peer-reviewed" papers?
    
    The CR-climate link may be complex and subtle, as Usoskin latest paper has shown: there is a regional pattern to cosmic rays. The effect on clouds depends on location, not just latitude. Until the effect is fully understood, parameterized, and included in GCM's, it seems a bit futile to try to refute it with simplistic arguments, and then claim that similarly simplistic arguments should not be used to refute AGW.
    
    Just my two cents...
    
    
    
    
24. Wondering Aloud at 04:51 AM on 30 April 2008
    I find interesting the claim that for 500 million years CO2 levels have correlated well with temperature! This only applies at all if you assume it is the temperature change causing the CO2 level change. I would be more convinced by your arguments if you could force yourself to leave this out of them. Past temperature swings cannot be explained by the relatively small CO2 level swings that follow them.
    
    Reminding us all that CO2 levels cannot explain past climate change is not a great way to convince us that it is the dominant force in current change.
    
    
25. shawnhet at 05:32 AM on 30 April 2008
    For clarification, my last post when I stated "earthquake" I meant to say volcanic eruptions. Oops! Sorry about any confusion.
26. chris at 06:58 AM on 30 April 2008
    shawnet,
    
    I don't think we're talking past each other. We both agree that the very marked warming of the last 30-odd years has had no significant solar contribution, including solar contributions to the cosmic ray flux (CRF). That's very straightforward, and no one disagrees with that. The evidence is rather strong. Svensmark, for example, makes the case for that interpretation, so one cannot assert that the "cosmic ray fluxers" are being "bullied" or ignored or whatever!
    
    Might the CRF have some contribution in the future? Might it have had some contribution in the past? Certainly it's possible. We don't rule anything out. So far, however, there isn't much evidence for a CRF contribution to the Earth's energy budget in the past.
    
    If we're concerned about the rather marked warming of the last 30-odd years, then we focus our attention to where the evidence lies. It's not the CRF obviously!
    
    Might the CRF have some effect on clouds? It's not very clear but the evidence is marginally consistent with the possibility, although the data isn't very compelling at all. After all we had a couple of papers in 2000 reporting a partial correlation....and then nothing but rather contrary evidence since..if the situation is so straightforward, why nothing to support the initial analyses?
    
    Might the CRF and its possible effects on clouds have some effect on the Earth's "energy budget"? We don't know. There isn't any evidence for such a possibility, although some have hypothesised such an effect. We await the evidence.
    
    Notice that no one disputes that there are many periods in the past of varying surface temperature that don't have any apparent role for the greenhouse effect. The mid-20th century temperature "stasis" was most likely due to atmospheric aerosols from dirty fuels....the warming of the early 20th century probably had a solar contribution (but there's no evidence for a role for the CRF for that event)...the cooling of the Little Ice Age very likely had a strong solar contribution; was the CRF involved? Who can say? There's no evidence for such a contribution...and so on....
    
    ....i.e., no one rejects the possibility of a potential contribution of the CRF to the Earth's Energy budget. There just isn't any evidence for it.
    
    As for your comments:
    
    "As to whether the majority of the studies support the connection, I will say that the majority that I have seen do so(as do the majority of the papers listed on this thread). We can obviously argue the merits of each individual paper, but I don't think this is a productive use of time."
    
    The majority of papers don't support the connection. A number of studies purport to support the connection, but they don't present compelling evidence, and in general follow up studies highlight flaws...even follow-up studies by some of the original advocates. For example Veizer has reinterpreted some of his paleotemperature data, such that the putative "correlation" between the supposed CRF and the Earth's paleotemperature is even less well supported by the evidence.
    
    You suggest that it's not a productive use of time to argue the merits of each individual paper. But if we're interested in the evidence that's surely what we should do.
    
27. chris at 07:32 AM on 30 April 2008
    Wondering Aloud,
    
    I'd like to address each of your comments in turn, since you've made a list of odd statements (apart from the first one!).
    
    ["I find interesting the claim that for 500 million years CO2 levels have correlated well with temperature!"}
    
    Absolutely. It is interesting and your exclamation mark is not misplaced!
    
    ["This only applies at all if you assume it is the temperature change causing the CO2 level change. I would be more convinced by your arguments if you could force yourself to leave this out of them. Past temperature swings cannot be explained by the relatively small CO2 level swings that follow them."]
    
    That's certainly incorrect and is a decidedly false premise. Why should we assume such a thing? After all we know that the warming at the Paleo-Eocene Thermal Maximum is likely due to massive release of greenhouse gases (methane likely predominant)...that the massive tectonic events associated with the formation of the Deccan Traps in now-India released large amounts of greenhouse gases around the time of the end-Cretaceous warming...a paper in last weeks issue of Science (April 25 2008) on the revised timing of argon-argon radiodating re-emphasises the likely role of the Serbian Traps eruptions in the Permian-Triassic extinctions....we know that long term weathering process can draw down atmospheric CO2 and result in cooling....that the earliest major glaciations in Earth's history may be explained by the destruction of atmospheric greenhouse methane from the rise of O2 in the late Archean...and so on...
    
    
    You are making a major logical error based, I suspect, on your understanding of the ice-age cycles. Obviously the primary driver of these processes was the achingly slow variations in the orbital properties of the Earth (Milankovitch cycles), and the greenhouse-induced warming was a feedback that enhanced the primary warming. But you make a big error in equating these either with the events in deeper paleohistory or even with the current warming. During the last glacial - interglacial transition, the atmospheric CO2 rose from around 180 ppm (glacial) to 280 ppm over 5000 years or so, in response to Milankovitch warming. That's a rather small amount produced at a slow rate (around 2 ppm per century averaged over the whole transition even if most of might have been produced during the earlier stages). We're raising atmospheric CO2 at more than 2 ppm per year now (100 times faster).
    
    And the past CO2 swings were certainly not "relatively small"! The warming at the Permian-Triassic boundary was associated with a rise of atmospheric CO2 to over 3000 ppm. Is that "relatively small"? Warming episodes in the Jurassic were associated with atmospheric CO2 levels that appear to have reached around 6000 ppm. "relatively small"? Warm periods in the Silurian and Devonian had atmospheric CO2 levels around 3000-4000 ppm...and so on. These are not "relatively small" changes. And the evidence indicates that the temperature changes followed these changes in atmospheric greenhouse gas levels. CO2 is a greenhouse gas after all, and when its atmospheric concentrations are increased a warming response is inevitable. That’s obvious surely…
    
28. chris at 07:46 AM on 30 April 2008
    Francois,
    
    on skepticism:
    Skepticism only has meaning with respect to some specific thing. This site seems to be about examining skepticism with respect to the science on global warming. Otherwise we’re all skeptics I would have thought, and we can apply our skepticism wherever we feel it appropriate to focus it.
    
    And then it comes down to the quality of the evidence.
    
    So addressing your point:
    
    [I find it funny that when it comes to cosmic rays, anything but a perfect correlation is taken as a refutation. But for greenhouse gases:
    
    "El Nino's, La Nina's, volcanic aerosols, the solar cycle and various stochastic elements of the climate system will all act to modulate the greenhouse enhanced temperature increase, sometimes countering this and sometimes adding to it. "]
    
    No one is asking for "perfect correlations", but we should be skeptical if the correlations don't exist or they seem to be somewhat contrived, or they correlate variables only over very short time periods, or they're reported 8 years ago, with no subsequent supporting data in the meantime, but only contrary data...and so on. We should be skeptical, for example, of assertions of cosmic ray flux contributions to the warming of the last 100 years. During the period when we've been able to assess the CRF (since around mid 1950's) there isn't any correlation, and 'though cosmic rays may have contributed to temperature changes/fluctuations during periods before direct measurement of the CRF was made, there's little evidence for this, and unless someone comes up with a compelling argument for including CRF effects in the Earth's "heat budget", there doesn't seem much scientific basis for asserting a significant role.
    
    Now you may argue that the fact that we haven't yet got much evidence for this contribution, and as a result we haven't got a good handle on its quantitation, doesn't negate the possibility of a significant potential CRF contribution. However we do know that the Earth can undergo a marked global scale warming under the influence of a rather large "heat budget" imbalance, without any solar contribution whatsoever, let alone a contribution from the CRF. So if we are assessing the cause of the large scale warming of the last 30-odd years and its implications, we don't pretend that factors that are demonstrably of little significance are not so. That doesn't mean that all sorts of solar contributions (CRF included) didn't make significant contributions in the past and won't in the future.
    
    All the other things "El Nino's, La Nina's, volcanic aerosols, the solar cycle and various stochastic elements of the climate system..." we know do make a contribution, if only transiently. We can measure these directly, even if we can't predict them (obviously we can predict the solar cycle, but not it's "strength"). We know that they result in transient modulation ("noise") of the Earth's equilibrium temperature (if the latter is at equilibrium), or of the transition to a new equilibrium temperature if there is a forcing that causes the Earth's "heat budget" to be in imbalance.
    
    We know that the Earth's "heat budget" is in imbalance. The Earth has warmed rather markedly over the last 30-odd years. This might have been due to an increase in solar output during this period. However we've measured this, and we know that it wasn’t.
    
    There may be an effect of the CRF on cloud formation. The evidence is inconclusive, although we're willing to accept that there might be an effect, and we await some compelling evidence.
    
    you say:
    
    "The CR-climate link may be complex and subtle..." Quite so. It's so subtle as to (so far) escape characterization by a considerable amount of scientific analyses.
29. shawnhet at 04:05 AM on 1 May 2008
    "The majority of papers don't support the connection. A number of studies purport to support the connection, but they don't present compelling evidence, and in general follow up studies highlight flaws...even follow-up studies by some of the original advocates. For example Veizer has reinterpreted some of his paleotemperature data, such that the putative "correlation" between the supposed CRF and the Earth's paleotemperature is even less well supported by the evidence.
    
    You suggest that it's not a productive use of time to argue the merits of each individual paper. But if we're interested in the evidence that's surely what we should do."
    
    I think it would be much more productive to simply list the scientific papers that demonstrate the CRF-cloudiness connection that you keep referring to. I am familiar with Sloan's(obviously) and Rahmstorf's, but I don't think either of these is all that good a critique. Do you have any others in mind?
    
    There are some results out there that are very difficult to dismiss (for instance Harrison and Stevenson 2006 and Svensmark's 2007 experiments) to uncritically accept the idea that most of the evidence is against such a link.
    
    Cheers, :)
30. chris at 09:52 AM on 1 May 2008
    shawnet,
    
    O.K. now perhaps we are talking past on another!. My very first post on this thread started:
    
    ["There are two questions really:
    
    (i) do cosmic rays influence clouds formation?
    (ii) is this effect, if real, significant with respect to the Earth's surface temperature and climates?"].
    
    When I state that most papers don't support the connection, I'm talking about (ii) namely the connection between CRF and climate. There is evidence that the CRF can influence cloud formation. Is this significant with respect to the Earth's surface temperature and climates? There's very little if any evidence in support of that idea.
    
    We know that the very marked warming of the last 30-odd years has occurred without a significant contribution from changes in solar outputs. Therefore the CRF has been largely irrelevant during that period. Svensmark has presented his evidence in support of that rather well-supported conclusion.
    
    Svensmark's ionization chamber experiments aren't particularly compelling with respect to this (CRF-clouds-climate), and a causal chain needs to address several questions:
    
    (i) are the nanometre size nuclei observed in the ionization chamber relevant for cloud nucleation in the real world?
    
    (ii) does any CRF-mediated cloud formation significantly alter the cloud cover in a climatically significant manner? After all there are many cloud nucleating species in our atmosphere (particularly in our modern world). Might any clouds produced by CRF be nucleated by other species anyway?
    
    (iii) Clouds have short lifetimes (few hours). The nucleation of clouds results in a lower water vapour concentration in that part of the atmosphere (unless or until the condensed water vapourises again). So a cloud that might subsequently have formed may not now do so. Thus the cloud cover in that particular region might be net-neutral averaged over a short period (say a daylight period) whether or not CRF-mediated cloud nucleation occurs...
    
    and so on. Note that this isn't quibbling. It is addressing the rather blatant facts that CRF hasn't contributed to the most significant global warming event of the entire Holocene, and the fact that there is no evidence for CRF-mediated cloud nucleating contributions to climate.
    
    Thus it is difficult not to be skeptical about the notion. Of course that doesn't mean that it doesn't exist (CRF-mediated cloud nucleating contributions to climate). It just means that there is no evidence for it..
31. shawnhet at 05:38 AM on 2 May 2008
    Chris,
    
    A couple of points here regarding the significance argument. No one that I am aware of has challenged Harrison & Stevenson 2006 which found that "Across the UK, on days of high cosmic ray flux (above 3600×102neutron countsh−1, which occur 87% of the time on average) compared with low cosmic ray flux, (i) the chance of an overcast day increases by (19±4) %, and (ii) the diffuse fraction increases by (2±0.3) %. During sudden transient reductions in cosmic rays (e.g. Forbush events), simultaneous decreases occur in the diffuse fraction. The diffuse radiation changes are, therefore, unambiguously due to cosmic rays. Although the statistically significant nonlinear cosmic ray effect is small, it will have a considerably larger aggregate effect on longer timescale (e.g. centennial) climate variations when day-to-day variability averages out."
    
    Now, it seems to me that given the above, it is one thing to be skeptical of the significance of the CRF climate link, it is another to say that there is no evidence that it is significant.
    
    Cheers, :)
    [ Response: Actually, I know of two papers that cast doubt on Harrison's result on forbush decreases. One is Sloan's paper, the subject of this post. His paper actually has 3 sections - on latitude dependence, forbush decrease events and LCC lagging CRF. My plan was to do a separate post on each section. However, just over the last few weeks, another study was done on forbush decreases by Jón Egill Kristjánsson who presented his results at a recent AGU conference. He found a similar result to Sloan. I contacted Kristjánsson about the paper but it's still in review. So I'm mulling over whether to wait for the paper to be published before doing the post on forbush events and Harrison's paper. Or maybe I'll just go with Sloan for now. Stay tuned :-) ]
32. chris at 21:48 PM on 2 May 2008
    Shawnet,
    I decided to have a good read of Harrison and Stephenson this morning, and have come up with as rather long post for which I apologize. I've tried to organize it coherently:
    
    Harrison and Stephenson [Proc. Roy. Soc. A 462, 1221–1233 (2006)] is fine. However I don’t think it provides very good evidence of a CRF contribution to climate. As the authors state “Changes in DF and the frequency of overcast days represent changes in the weather and the atmospheric energy balance.”, and there are a number of considerations that need to be addressed if the effect is to be considered a “climate”-influence, rather than a “weather”-influence.
    
    Here are some:
    
    (i) The first relates to something that we are all aware of. There isn’t any correlation between these processes and the Earth’s temperature evolution during the period in which the CRF has been monitored in detail. After all the CRF was as high in the period 1950-55 as it is now. High CRF should correlate with high low level cloud and low temperatures. However the Earth is clearly a lot warmer now than in the early 1950’s.
    
    Now of course, that’s an unfair comparison, since the CRF varies with the solar cycle, and so one doesn’t expect persistent warming/cooling from an oscillating forcing that has no trend. But that just re-emphasises that the there has been no significant solar contribution (CRF included) to the very marked warming of the last 30-odd years. This has been re-emphasised in two detailed analyses just published by Lockwood/Frolich (see bottom of post [***]).
    
    
    (ii) Harrison and Stephenson make two points in their paper. If one looks at their Figure 2 and associated text, one sees [point ONE] a very weak correlation between the CRF and the diffuse fraction (DF) (cloudiness – presumably dominated by changes in low-level cloudiness), which [point TWO] breaks down when the CRF is higher than 3600 (x100) per hour. Above this level the DF is independent of the CRF.
    
    
    Taken at face value, this seems problematic. Examination of Harrison and Stephenson’s Figure 2 indicates that most of the data in their analysis period (1968-1994) lies in this non-responsive region of the CRF/DF “correlation” where the CRF and DF are apparently completely independent (no correlation). If one examines the entire CRF record [http://ulysses.sr.unh.edu/NeutronMonitor/Misc/neutron2.html], that is reinforced. During around 90% of the entire period between the start of the CRF count (1951) to now, the CRF has been higher than 3600 (x100) per hour [if one looks at the climax CRF/sunspot number plot on the page I urled and compares this with the data in the downloadable datasets on that page one can work out that a CRF count of (3600 (x100) per hour corresponds to around the 83% level].Again one might conclude that that’s additional explanation for the complete absence of a detectable CRF (solar) contribution to the marked warming of recent decades.
    
    
    (iii) This leads to an additional problem (it seems to me), which relates to ascribing CRF climate effects to earlier periods. Although we’re in a solar minimum right now with respect to the solar cycle, the sun is in a relatively “strong” state. If one examines the sunspot data going back to 1600 [e.g. here: http://en.wikipedia.org/wiki/Sunspot], this is apparent. Now if the very strong inverse relationship between CRF and sunspot number (see Climax neutron/sunspot data urled above) was maintained in those periods (no reason to think otherwise), then the CRF was likely generally stronger [greater than 3600 (x100) per hour] during periods like the Maunder and Dalton minima. However taking Harrison and Stephenson’s data at face value, the cloudiness as measured by the DF is completely independent of the CRF at values above 3600 (x100) per hour. Therefore the CRF cannot have played a significant role in the cooling during those periods.
    
    
    We could deal with that problem by making some ad hoc assumptions. Perhaps the CRF became decoupled from the other solar parameters during periods with “weak” solar output. Perhaps the thresholds for a CRF-cloud correlation was higher then than it has been for the past 40 years (one might imagine that in periods where the air is rather clear of man made aerosols, the threshold for CRF effects on cloud formation become higher…and so on). But there’s no evidence for any of these ad hoc assumptions.
    
    
    (iv) One may still question whether the apparent correlation between CRF and DF in the CRF range below 3600 (x100) per hour is due to the CRF and not some other solar-related parameter that cycles in step with the CRF. Harrison and Stephenson used Forbush events in an attempt to assign a causal connection. However the useable Forbush events are apparently very rare (31 days in a 50 year period with CRF count reduction greater than 5%), and so the data is not terribly strong; the Forbush events (0.2% of the record) may not be representative of the other 99.8% of the record.
    
    
    (iv) We come back to the point of whether the CRF-cloud “correlations” are significant with respect to “climate” effects as opposed to “weather” effects. Sloane and Wolfendale’s analysis indicates a lack of detectable temperature effects from the Forbush events. We’ve already seen [see point (i)] that there’s no correlation between the CRF and the Earth’s temperature response since the start of the count, although we recognise that the CRF has been trendless, and so will only be represented in the solar cycle which is difficult to “pick out” in the record since it is smoothed by the inertia in the climate system, and is apparently small with respect to other stochastic variations (El Nino’s, volcanos, La Nina’s), and especially with respect to the strong and persistent trend arising from the enhanced greenhouse effect.
    
    
    (vi) A couple of other points need to be established if the CRF-cloud correlations have potential “cloud” implications rather than “weather” implications:
    
    (a) Does CRF-mediated cloud formation result in a nett change in the cloudiness over that resulting from other cloud-forming process (currents that mix warm damp air with cold air; all the other atmospheric aerosols; mountains and land masses). I don't think that's been established 'though Harrison and Stephenson's data might support that conclusion?
    
    (b) What about the day/night relationships? According to the CRF-cloud theory, cloud effects act during the day. CRF-mediated cloud formation blocks solar irradiance at the surface cooling the Earth. Notice that the greenhouse-mediated warming acts during both day and night, since the effect results from atmospheric “trapping” of IR radiation emitted from the Earth’s surface that occurs during the day and night. This is consistent with the fact that the warming of recent decades is as large or larger for night-time than day-time periods, and is another inconsistency with any potential CRF contribution to recent warming. One possibility is that any CRF-mediated cloud effects roughly cancel between day and night. For cloud effects to act on climate (Earth’s surface temperature), one expects that they need to be persistent at least for some hours. However enhanced cloud formation during the day (cooling effect) that persist into the night will result in a night time warming effect (decreased radiation of surface warmth) and vice versa.
    
    Overall Harrison and Stephenson is an interesting paper with what looks like fine data. However it doesn’t negate the very strong evidence for a lack of CRF effect on the very marked warming of recent decades, and if we’re concerned about that, we should focus to rather more well-supported causal factors. Secondly, it doesn’t constitute a very significant case for a CRF-climate link, and in fact tends to dis-favour such a link, at least over time periods in the relatively recent past (since 1600) when there is a reasonably good record of solar outputs (as monitored by sunspot numbers).
    
    
    
    [***]M. Lockwood & C. Fröhlich (June 2008) Recent oppositely directed trends in solar climate forcings and the global mean surface air temperature. II. Different reconstructions of the total solar irradiance variation and dependence on response time scale Proc. Roy. Soc. 464, 1367-1385.
    
    M. Lockwood (June 2008) “Recent changes in solar outputs and the global mean surface temperature. III. Analysis of contributions to global mean air surface temperature rise” Proc. Roy. Soc. 464, 1387–1404.
    
    
33. chris at 23:23 PM on 2 May 2008
    A slight clarification of my long post. When I say (referring to Harrison and Stephenson's data; e.g. see their Figure 2a) that above a CRF count of 3600 (x100) per hour there is no correlation between the CRF and the diffuse fraction (DF), I really should say no correlation between changes in CRF and changes in DF. Above the 3600 (x100) per hour "threshold", the DF remains high, but loses its correlation with the CRF that is marginally apparent at CRF values below the threshold.
    
    And similarly, in pointing out that periods in recent history (last 400 years as indicated by the well-characterised sunspot numbers) like the Maunder and Dalton minima can't have had a contribution from the CRF since the CRF was likely well above the 3600 (x100) per hour threshold during these and other large parts of the last 400 years, I really should say that differences between the surface temperature of the Maunder or Dalton minima (say) and the surface temperature during the period between 1975 and 1925 (say) can't have had a significant contribution from any changes in the CRF, since it's likely that the CRF was already above the threshold value [3600 (x100) per hour)] where changes in the CRF don't result in further changes in the DF (cloudiness), during all of these periods.
    
    and in point (iv) I meant "persistent cloud effects" rather than "temperature effects" in realtion to the Sloane/Wolfendale analysis of Forbush events!
    
34. shawnhet at 03:03 AM on 3 May 2008
    Chris, I haven't read H&S in a while, but I will make a couple of points.
    
    "(ii) Harrison and Stephenson make two points in their paper. If one looks at their Figure 2 and associated text, one sees [point ONE] a very weak correlation between the CRF and the diffuse fraction (DF) (cloudiness – presumably dominated by changes in low-level cloudiness), which [point TWO] breaks down when the CRF is higher than 3600 (x100) per hour. Above this level the DF is independent of the CRF."
    
    I will agree that this seems to be a pretty good point against a simple relationship btw CRF and cloudiness. I will have to read H&S again to, and think about it, however, the relationship found by H&S is still statistically significant and is consistent in multiple locations. IAC, I think it is a mistake to claim that the relationship btw CRF and clouds is simple, but it is also a mistake to claim that there is no evidence of such a relationship.
    
    Cheers, :)
    
35. m_b at 23:47 PM on 25 August 2008
    Terry Sloan is careful to make clear that they have never said that there is no connection between cosmic rays and cloud formation. He just believes it's not the whole story when explaining cloud formation.
    
    Even Sloan believes the strong negative correlation with historical temperature reconstruction's vs radionuclides and radiocarbon in ice cores is interesting and requires further investigation.
    
    Personally I can't escape the logic that these beryllium-10 and radiocarbon ice deposits (formed from GCR interactions in our atmosphere), indicate that GCR's either directly, or indirectly have some effect, or show an unknown effect on our climate. That GCR's come from outside our planet, dictates logically to me that climate is strongly influenced by external factors.
    
    When I look at the massive sudden drops, and sudden increases in temperature from the GISP2 Greenland Ice Core data at the start, and the end of the Younger Dryas (only approx 12,000 years ago), I am left speechless... these huge changes in temperature take place in mere decades, the extreme being a 5 deg change in just three years!
    
    I certainly need something better to explain these sudden cooling events, and more importantly, the sudden warming events, than Co2.
    
    On a side issue, periods in our past showing these large Be-10 levels, formed from CR's, are increasingly being pointed to as a possible explanation for massive globalised Gene duplication events appearing at the same time, and postulated to be one of (if not the main) driver of evolution.
    
    

Post a Comment

Foul language, trolling, personal attacks or non-relevant links will be deleted.

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