Why I am still Skeptical about Sunspots and Temperature


The fact that the latest sunspot cycle is very weak has gotten lots of news lately.  Many are claiming that a lack of sunspots will cause the Earth to cool in the near future.  I am going to stand alone as a skeptic and say I am not convinced.  The data simply does not support that conclusion.  I am going to explain why I do not believe the prediction that low solar activity (as measured by low sunspot counts) will cause a reversal in the warming trend.

I am going to start off with the basic chart showing the average monthly sunspot count since 1749.

The Inconvenient Skeptic

(Orange) Average Monthly Sunspot count, (Black) 11-yr moving average of the count.

Timing is everything.  The timing of the sunspots is critical for this analysis.  The best theory for sunspots changing the climate is that higher sunspots prevent cosmic rays from nucleating cloud formation.  So an increase in sunspots will reduce cloudiness which will increase temperature.  A reduction in sunspots will increase cloudiness which will decrease temperature.

Here is the HadCRU3 temperature since 1850.

The Inconvenient Skeptic

HadCRU3 Global Temperature

What isn’t clear is the problems with the timing that are present.  Certainly sunspots in the past 60 years have been higher, but the sunspots were highest in 1960 and the temperature has been highest in the past decade.  Nor does a significant time delay explain the problem.  The temperature simply does not appear to be overly dependent on the sunspot count.

The Inconvenient Skeptic

(Green) 3-yr moving HadCRU3 Anomaly, (Orange) Monthly Sunspot count.

Since 1993 the sunspot count has not been significant.  Certainly in the past 18 years there has been no reason from sunspots to explain the elevated temperatures.  Even before then the timing does not indicate a cause and effect relationship.  I am going to zoom in on a 100 year period from 1880-1980.

The Inconvenient Skeptic

100 years of sunspot and temperature anomaly.

Until the very end of the period, the highest temperature took place in the 1940’s.  Sunspots did increase prior to temperature there, but 1917 sunspot peak was comparable to the 1937 peak, but the temperature was not.  After the peak temperature in the 1940’s, the temperature dropped for the next 30 years while that same 30 year period experienced the highest sunspot activity that has ever been recorded.

In a very general sense the temperature of the Earth has been higher lately and sunspot activity has been higher over the past 70 years, but there is no clear indication that sunspot activity has a predictable influence on the Earth’s temperature.  That the Earth was cold during the Maunder Minimum, but all indicators are that it was even colder during the 1500’s when solar activity was not low like it was in the 1600’s.

There has been on big solar activity reconstruction by Solanki in 2004.  I have also compared that to the Moberg reconstruction from 2005.  Once again it is very difficult to see the influence of solar activity on the Earth’s temperature.

The Inconvenient Skeptic

(Blue) Moberg 2005, Temperature Anomaly. (Orange) Solanki 2004, Solar Activity Reconstruction

While it does show very low activity 500 years ago when temperatures were cold, it also shows relatively low activity 1,000 years ago when temperatures were higher than normal.

So in the modern day high resolution data there is not clear effect of solar activity on the Earth’s temperature.   In the long term reconstruction there is also a distinct lack of solar activity causing changes to the Earth’s temperature.

Based on the available data it is impossible to make any useful temperature predictions based on solar activity.  To do so is the same exact trap that warmists have fallen into with their conclusion that CO2 levels have caused the recent warming.  There is no difference in the two conclusions.  Very broad correlations like solar activity and CO2 levels do not prove anything.   While Solanki concluded that the solar activity over the past 70 years is “exceptional” the actual temperature of the past 70 years has not been exceptional on the scale of the Holocene.

While I cannot say that there is no influence at all, it is clear that it is not a driving force of the Earth’s temperature and the prediction that low solar activity will reverse warming is unwarranted.   The Earth will cool down in the future and it might happen 5 years from now, but it might not happen for another 30 years.   Anyone who predicts that the low solar activity will cause cooling risks appearing as foolish as those the predict warming based on CO2 levels.

Posted in Climate and Science Overviews by inconvenientskeptic on July 16th, 2011 at 8:20 pm.

15 comments

This post has 15 comments

  1. Richard111 Jul 17th 2011

    Okay. I’m going to stick my neck out here as I am not qualified to discuss this at all. My intuition is that the global long term temperature is controlled by the the mass of the oceans.
    It is easy enough to calculate the mass and specific heat of a column of the atmosphere on a one square metre base. It then appears just 2.6 metres of ocean under that column has the same heat capacity. There is simply no contest between the heat capacity of the total mass of the oceans against the heat capacity of the atmosphere. Also the ability of the atmosphere to lose heat energy to space compared to the oceans losing heat to the atmosphere is again a no contest exercise.
    I believe air temperatures are of interest to the specific locations but otherwise meaningless in the long term.
    By long term it would seem something like 800 hundred years and more before you can be sure of a positive change in global climate.

  2. My opinion is that it’s too early to draw conclusions. I have a couple of problems with your analysis here. Sunspot data corresponds pretty well with major climate shifts when we are talking about actual observed sunspots. In your graph of 2,000 years of temperature and sunspots most of the graph is covering time periods before the invention of the telescope, so the sunspot estimates prior to 1600 are conjecture based on some kind of proxy. Is the data shown after 1600 also based on the same proxy or is it actual observations?

    Second your graphs using more recent data start in 1850 and 1880, both of which are well past the Mandauer and Dalton minimums. A proper assessment of this issue should include the period between 1650 and 1850, and this has been done by others. The correlation (and at this point it is only a correlation) between low sunspot numbers and very cold climate is very strong. I would encourage looking up ground temperature records from the UK and comparing them to sunspot number – I believe the record for the UK goes back to 1650 and is reasonably reliable. Sunspot observations from the period are known to be reliable as well, so it would be a reasonable exercise.

    In any case an attempt to evaluate this issue properly is being carried out at CERN. Watch an interesting presentation here:

    http://cdsweb.cern.ch/record/1181073

    Also we’ll find out soon enough if the climate cools in the near future in response to reduced solar activity.

  3. Greg2213 Jul 17th 2011

    You make a great point. Maybe after another couple hundred years of good observations we’ll have a better idea of how solar variations drive or don’t drive climate changes.

    Obviously increasing CO2 (from current levels) is a bit player in things. We can perhaps rule out solar variations, but there clearly something cyclic driving the climate system and the several climate cycles that we can observe. Clearly the sun provides the energy to drive a livable climate on this planet. So if it isn’t solar variation, what is it? Random/chaotic “walks” seem unsatisfying.

    I get the strong impression the we really don’t understand what drives the climate system, at all.

  4. Well its likely that all of these factors play a role. Its not ocean cycles OR sun spots. At times, like when sun spots drop to zero for a long period maybe the sun dominates. Over the past century when the variation in sun spot peaks and minimums was not so great cycle to cycle, then ocean circulations like PDO are more dominant. Maybe sometimes they come together – an extremely low sunspot cycle + negative PDO creates a tipping point toward a really cool climate period. Recently I saw an interview with a German scientist (sorry don’t recall his name) but he said Co2 accounted for about 10% of climate variation. I think that is a reasonable stance to take, so yes we understand Co2 is a greenhouse gas (not cartoon deniers) but the question is does Co2 play the role as a significant or the dominant driver of climate (no).

    I agree with Greg, the sun provides the energy to drive a livable climate on earth. Oceans do not provide energy although the store and move heat. Ultimately the oceans though get energy from the sun.

  5. inconvenientskeptic Jul 17th 2011

    Sunspots did not appear to be significant so they are not there. Since they had been in the news I wanted to explain my conclusion on sunspots.

    The oceans are important in controlling climate because ocean currents can keep polar regions far more temperate. Anything that blocks the flow of energy carried by ocean currents will impact climate.

    While I consider the things like the AMO and PDO to be important for short term variation, it is the longer term variation that is more significant. Their role in that is far more limited.

    In the end it is the relationship with the sun that controls
    climate. I spend a fair amount of time in the book explaining that rather complex relationship.

  6. Charles Higley Jul 17th 2011

    The problem with your discussion is that your temperature graph is erroneous. The 1930s peak was warmer than the recent peak. Contamination of the temperature records for recent times has bloated the temperatures. We have the urban heat island effect (which is poorly, if at all, corrected for), drop out of rural sites (casuing a warming of the data, poor maintenance of many sites (obstructions lead to heat retention), and simple dishonest addition of value to recent data and even decreases to past data.

    There is also the really wrong strategy of extrapolating huge areas of the planet from sites ridiculously far away. In one case part of the Andes is given the temperature of the Amazon basin. Then we have temperature data from sties that fail being cobbled up from splicing data from other nearby sites that are not in the same situation (different altitude and distance from the ocean). It is simply swrong toextrapolate the Artic region from sites around the rim which are always in towns or near airports.

    Corrections are not made for these problems and the data is altered because they want to show warming for the recent years. They give some lip service to corrections, but it’s not real.

    So, you should have a problem when you are trying to understand any relationship between the real and the adulterated.

  7. inconvenientskeptic Jul 17th 2011

    Charles,

    If I assumed warmer 1930’s than today then there is even less support for the theory that sunspots cause warming.

    While there are problems with the temperature data, taking those issues into consideration only further weaken the sunspot theory of warming so there is no impact on my analysis.

  8. So, are you saying there was no correlation between sunspots and the cold of the little ice age, or the year without a summer? Also Willie Soon of the Harvard Smithsonian Center for Astrophysics shows a very strong correlation between sunspots and arctic air temperatures. See figure 1 in :

    http://www.cfa.harvard.edu/~wsoon/myownPapers-d/Soon05-SolarArcticTempGRLfinal.pdf

  9. inconvenientskeptic Jul 17th 2011

    The reference is annualized total solar irradiance and not sunspots. Then the temperature is arctic only.

    I am not saying that there is no correlation, but in too many instances both short and long term it is not consistent enough to make accurate predictions.

    Little ice age seems a good fit until you use that as a baseline for the MWP. Consistent and predictable response is the standard for cause and effect to be determined. Sunspots do not meet that criterea.

  10. inconvenientskeptic Jul 17th 2011

    I have to be fair and hold sunspots to the same standard I hold CO2 too. While correlation does exist at times, it does not meet the cause and effect standard.

  11. If you were using two separate linear models to represent the relationship between two non-linear cyclical and some-what chaotic natural processes of unequal proportions with many uncertain dimensional variables and interactions, and they looked like they could fit together to show an obvious cause n’ effect, wouldn’t that be more surprising?

  12. You may be wrong. The correlation shown in the article below between cosmic ray flux and temperature is very strong, and its known that sun spots are an indirect indicator of the modulation of cosmic ray flux. This doesn’t offer proof, but it makes a strong case that there is a large influence and it should be investigated.

    More interesting in the article though, is that the chief of CERN has ordered physicists not to interpret the results of the CLOUD experiment. That is a stunning thing to do. The problem is the results of the experiment contradict AGW theory I suppose.

    http://www.theregister.co.uk/2011/07/18/cern_cosmic_ray_gag/

  13. inconvenientskeptic Jul 18th 2011

    If the assumption is made that cosmic rays strongly influence climate, but the sunspot data is what it is, then the conclusion would be that sunspots sometimes influence cosmic rays more than other times.

    I am not saying there is no influence, but I am saying that sunspots are not a reliable predictor of climate. If cosmic rays are, then the variable is the influence of sunspots on cosmic rays.

  14. SteveF Jul 23rd 2011

    Greetings,
    Firstly like your site. A reasoned voice.
    Secondly, The relationship between solar activity and sunspot activity is a little nebulous, but more indicative than not. A better indication of incoming (solar?) radiation is given by the production of Beryllium10 and Carbon14 (not used after atmospheric nuclear blasts). These isotopes are produced in the atmosphere by the interaction of nitrogen and oxygen atoms with “radiation” from both solar and “galactic” sources. The Sun’s effect dominating (?). I have a feeling the Solanki 2004 paper used Be10 (?) as a proxy for Sunspots.? (Dave ).

  15. inconvenientskeptic Jul 24th 2011

    Steve,

    You raise an interesting point with the statement about C14. If there is one thing that humanity has managed to mess up is proxies. Many, many things that had delicate balances are now out of whack.

    Of course messing up the proxy concentration does not mean that the attribute that was measured by the proxy will shift. Atmospheric nuclear detonations increased C14, but all that means is C14 is no longer a useful proxy.

    Warmists have confused the (small) warming effect of CO2 with the proxy effect that CO2 has as a result of temperature dependent ocean solubility.

    One thing we have done is ruined CO2 as a proxy for ocean temperature. But that is all we have really done.

Web Design & Dev by

Mazal Simantov Digital Creativity