by Gavin Schmidt and Michael Mann
Solar forcing of climate is a subject that gets far more attention than any new observations or improved understanding would warrant. Two new articles appearing today attest to that. One is occasionally tempted to agree with Oscar Wilde when he said:
Why does not science, instead of troubling itself about sunspots, which nobody ever saw, or, if they did, ought not to speak about; why does not science busy itself with drainage and sanitary engineering?
Except in this case it’s Nature rather than Science who are troubling themselves…
A new review paper by Foukal et al does a reasonable job summarising the mainstream opinion on the issue. In particular, they outline quite clearly why some ideas related to long term solar variability (such as solar disk radius changes, or the difference between cycling and non-cycling stars) have recently fallen out of favor. Indeed, they assert that there is little evidence for any solar variability in irradiance that is not related to the shielding/enhancements of sunspots and faculae – which implies only a modest decrease in solar flux at the Maunder Minimum for instance. We could quibble with their use of paleo-reconstructions, their climate modeling approach, and the rather cursory treatment of the substantial body of work relating to amplyfying mechanisms due to UV/ozone links, but we’ve gone over this ground before and we refer readers to those earlier discussions.
Slightly more novel is the description in the News section of a new experiment at CERN that is attempting to test the cosmic-ray/cloud hypothesis by building a large cloud chamber mimicing the atmosphere and firing high energy particles at it. There’s nothing intrinsically wrong with the idea, but given the $11 million to spend on climate change research, we would have tended to favour projects that, unlike this one, have at least some empirical support within the observations….
Re #44
“If those (1,470-year) cycles persist across both glacial and interglacial climates, it is hard to argue that they are the result of multidecadal or multi-centenial internal climate modes rather than some external forcing.”
The Braun et al. paper clearly states that these cycles do not occur in interglacial climates like the Holocene, neither in models nor in observations. Why arguing with it then?
Re #46
Besides the problems of the link to the 22-year and lower frequency cycles in the Scafetta and West paper (see https://www.realclimate.org/index.php/archives/2006/03/solar-variability-statistics-vs-physics-2nd-round) you have to be aware that the solar influence during the last two solar cycles which is given by Scafetta and West is based solely on the TSI (total solar irradiance) composite of Willson and Mordinov, which shows a positive Trend of TSI (+0.045% from the solar minima 1986 to 1996). However, there are two independent other TSI data sets, by Frohlich and Lean and by Dewitte et al., which show no significant TSI trend over the last two cycles (slightly negative and +0.01%, respectively, both non-significant). If you use these trends in the Scafetta and West calculation, the solar influence is very near to zero.
The trend found by Willson and Mordinov in contrast to the other two composites seems to be an artifact for two reasons:
a) the positive trend is not due to a long-term increase but the result of a short episode of increase (1989-1992) found in the data of one satellite (Nimbus 7). This increase has not been measured by the other satellite measuring at this period (ERBS);
b) other indicators of solar activity, which are closely correlated to TSI (sunspot number, faculae, geomagnetic activity) show no trend in that period.
Thus it seems very likely that the solar influence on global warming of the last two decades found by Scafetta and West is based on an artifact in the Willson and Mordinov composite.
References:
Frohlich, C., and J. Lean, 2004: Solar radiative output and its variability: Evidence and mechanisms. Astronomy and Astrophysics Review, 12, 273-320.
Dewitte S., D. Crommelynck, S. Mekaoui, and A. Joukoff, 2005: Measurement and uncertainty of the long-term total solar irradiance trend. Solar Physics, 224, 209-216.
Willson R.C. and A.V. Mordinov, 2003: Secular total solar irradiance trend during solar cycles 21 and 22. Geophys. Res. Let., 30, 1199-1202.
Re #48
Grant, if I correct the detrended global temperature of the last 30 years for the influences of ENSO (for the correction I took the MEI-Index*0.1 with a time lag of 5 months) and correct for the cooling of the volcanoes El Chichon and Pinatubo) I end up with a signal that shows a good correlation to the 11-year cycle TSI solar signal. Other studies with more sophisticated regression analysis have found a similar result (see my comment 5 above). Thus there seems to be a signal of the 11-year cycle. However, I can’t find any temperature signal of the 22-year cycle. At least it would be much smaller than the 11-year signal.
Re #52%53:
Urs, I have read the comment of Lean on the Scafetti and West paper, and their reply to Lean’s comment (on the AGU pay-per-article pages…).
I am inclined to agree that the Willson and Mordinov ACRIM satellite composite is probably an artifact. But at the other side, if there is a small shift, Scafetti and West are right that a shift over the 22 year cycle has more effect (app. 1.5 times) than an equal shift over an 11 year cycle. Shifts over longer time spans (like the 1900-1945 period) of course would induce even more warming and may have some (residual) effect even today.
Btw, there is no change in solar activity deducable from sunspots, as these are (near) zero anyway at solar minimum. At the other hand, there is a shift of ~1% in low cloud cover between the solar minima of 1986 and 1996 in Fig.1 of Kristjansson ea.. As there is a significant negative correlation between low cloud cover and solar irradiation (TOA), this may be an indication of a higher solar minimum in 1996…
I am not sure if this is applicable but there appears to be a solar pulse that has a 1500 years cycle commented on by Fred Pearce in his book “the last generation”. It appears to have empirical backing according to the chapter in the book from alleged reputable sources. In fact the little ice age etc seems to be a European phenomenon only rather than a world wide one like AGW is.
I would state than the Sun in not a major contributor to climate change if at all world wide. Sounds like the skeptics once again grasping at straws to me
Re #54 ‘shifts over a longer time span…may have some residual effect even today’
-I was wondering how long it takes for the global temperature to settle after, say, a permanent 1 W/m2 increase in insolation and whether a reduction of 1 W/m2 should take exactly the same time to reach equilibrium?
In the back of my mind I was wondering whether the cycling of the insolation (in itself) leads to a degree of ‘pumping’ ??
[Response: You can see that in the ‘committed climate change’ runs that were done for IPCC AR4 and reported in Meehl et al, 2005 for instance. Temperatures slowly rise (depending on how quickly the oceans warm) and give you the majority of the equilibirum signal in surface temperatures after a few decades. Sea level rise takes much longer to equilibriate. – gavin]
Re #50: Jon, unfortunately this is a rather ill-defined question. The Kyoto Protocol specifies average emissions targets for certain nations (basically, the industrialized ones) for the 5-year period from 2008 to 2012. To figure out how much effect this would have, you have to make assumptions about what happens after that period, what the emissions are like for other countries that don’t have targets, etc.
Look at it this way: Even if the whole world didn’t emit any greenhouse gases for the 5-year period from 2008 to 2012 (we all went on holiday to Mars?) and then went back to our emitting ways in 2013, we would only have delayed reaching a given level of CO2, and thus warming, by about 5 years. The point of this exercise is to show you that you should take any statements about how miniscule the effect of Kyoto is with a grain of salt.
In reality, Kyoto is part of a process. The main objectives of that process are (1) to start stabilizing the greenhouse gas emissions levels as a first step toward starting to reduce them and (2) to put a cost on CO2 emissions so that the technologies are developed and implemented to further reduce them. Because, despite what some folks will have you believe, markets don’t magically solve problems that they don’t know about…and technology is unlikely to save us as long as there is no financial incentive to develop and implement that technology. And, as long as one can use the atmosphere as a free sewer, that financial incentive simply does not exist.
So, in short, all those statements about how little Kyoto will do are just attempts to redirect your attention away from what the real issues and purposes are. Of course, one can have legitimate discussions about the best way to go about tackling the problem of climate change and whether Kyoto fits the bill…but I don’t think a good way to do this is by trying to estimate the long-term impact of a treaty that regulates emissions over only a short period, for only some countries, is only a first step, and is meant not so much to reduce emissions by a specific amount as it is to create the economic incentives for the development and implementation of the technologies to reduce emissions. Or, at least if one does make such an estimate, one has to very clearly state the assumptions one is making because the answer you get will be almost entirely dependent on those assumptions. Needless to say, I have almost never seen any of those statements about the miniscule effect Kyoto would have being accompanied by the any explanation of the assumptions whatsoever, let alone a sufficient explanation.
#54 re-greetings Ferdinand, I am interested to see if I understood what you said, there is a 1% increase in low cloud coverage during solar minimas? I am interested in the relation between clouds and heat, intuitively there are less clouds in very warm or very cold air.
Re 54
Ferdinand, while I can find a signal of the 11-year cycle in the temperature data, I can’t see any apparent 22-year signal. Neither in the temperature, nor in the TSI data (if you can, please show me). The amplitude must be much smaller than the 11-year signal.(I can’t reproduce the amplitude of the 22-year cycle presented by Scafetta and West). However, even if the effect of a 22-year signal is 1.5 times stronger: if there is no trend in TSI there is no effect on the temperature trend.
If we agree that there is almost no change in TSI (neither in maximum nor in minimum) and that there is a correlation of TSI and low cloud cover, the fact that low cloud cover is decreasing would lead me to the conclusion, that this decrease is likely produced by another factor than TSI (circulation, GHG, etc.).
Not to forget that there might be artificial trends in the ISCCP low cloud data … (see
Norris J., 2000: What can cloud observations tell us about climate variability? Space Science Reviews, 94, 375-380.)
Re 47:
Keep in mind that the younger dryas and heinrich events are regionally variable, with the strongest effect in the North Atlantic, and much less impact on the southern hemisphere.
Re: #51 Urs Neu
I wasn’t referring to the 1470 year cycle but to the multi-decadal Gleisberg and Suess cycles, which are detected in the interglacials. However, if they exist, presumably their 1470 year superposition would then also exist in the current interglacial but for some reason without being as apparent in the interglacial proxies as they seem to be in the ice ages.
Re #58,
Wayne, some time ago… I have looked at the (GISS global land+sea) temperatures of 1986 and 1996 (the subsequent minima in the satellite era), the difference in temperature (yearly average) is about 0.17 K, the difference for the 5-year average is 0.2 K. Thus it seems that either the increase in SAT (SST?) decreases low cloud cover with ~1%, or the low cloud cover change increases SAT/SST…
Re #59,
Urs, agreed, if there is no trend, then there is no effect… S&W nevertheless could produce a 22-year cycle beyond the 11-year cycle, based on the Lean ea. solar reconstruction 1900-1980…
The change in radiation balance (+ 1.4 W/m2 between the 1980s and 1990s for the 20N-20S equatorial band, according to the revised ERBS satellite data) is much too large to be caused by GHGs. The increase in GHGs in the same period induces some order of magnitude lower extra forcing. Except if this represents a tenfold positive feedback. But that would have led to a Venus effect, some long time ago…
Indeed, long-term cloud cover and radiation balances still lack the accuracy which is needed to make them reliable enough for climate understanding…
#62 Ferdinand, I agree that there should be a decrease in low cloud coverage with higher temperatures, it makes sense, although .17 C is an average , it means likely that some places in the world had much higher temperature increases.
I might have read it right,
during winter there are more clouds covering the Earth NH than summer, it would be nice
if it was a truism year after year, but I have not read much results on likewise cloud coverage research.
RE #54 Urs, I popped the Lean TSI and the HadCRUT3v series into PAST (a freeware educational stats package).
With the CRU temperature data this gave spectral peaks at around 21 and 9 years. With the TSI data it gave only the 11-year cycle (there was a very slight 22yr molehill.)
However I think that S&W are conjecturing that as well the direct (irradiance) effect there could be an indirect (exotic physics) effect that follows the 22-year reversal cycle. (Which is why they sliced rather than trended the ACRIM data).
I wonder if the 22-year temperature component is unique to the CRU dataset?
Re: solar cycle in temperature time series
I’ve looked again at the HADCRU temperature time series, just to be sure. I too see the peaks at periods 9.2 yr. and 20.3 yr. Furthermore, their statistical significance levels computed in the usual way are quite high — undeniably significant. The 9.2-yr. period, for example, yields a “chi-square per degree of freedom” statistic of 31.8, quite beyond any doubt. I’m using the “Cleanest Fourier Spectrum,” described here.
The problem is that the temperature time series shows serial correlation (as has been discussed here often). This means that statistical significance measures will be artificially inflated. One way to handle this for Fourier transforms was outlined in this paper (quite technical). The gist is that for data with serial correlation, you can’t compare the statistical measure to any set critical value, rather it should be compared to the rest of the spectrum to evaluate significance.
Comparing the aforementioned peaks (periods 9.2yr and 20yr) to the rest of the spectrum shows that they don’t quite make the statistical significance level (the 9.2yr period just misses by a hair). If we were to use the standard test, comparing to a set critical value, there are at least 27 peaks in the spectrum which have statistical significance! I don’t think any of us believes that the HADCRU temperature time series has 27 or more genuine periodicities.
So, it still seems to me that the presence of a solar-cycle response in the global average temperature time series is not yet established with statistical signficance. That’s not to say there isn’t one; it’s just not strong enough to be firmly established — yet.
Re: Grant #66
Thanx for your informative discussion. However, why should the null hypothesis be that there is no solar cycle response in the temperature? The sun is the dominant forcing in our climate system, and the temperature responses to its cycles are what they are. (profound?) The signal that we see is plausible on physical grounds. It may be a bit of a mystery why the response signal is not stronger, but the default should not be to dismiss it.
Those that want to dismiss the signal have a burden of proof, just as those that want to make more of it than appears justified, also have a burden of proof. Still it is valuable to know what its level of statistical significance is based on the various null hypotheses.
Ah yes here we are according to the book I mentioned in an earlier post.
The little Ice Age in Europe along with the so called medieval warm period and the like seems to be attributed by climatologists to the Sun.
Gerard Bond a researcher at the Lamont-Doherty earth observatory of columbia university proposes the idea of a pulse in the climate system which seemingly has a pulse of 1500 to 8000 years or so. The evidence for this comes from Heinrich events and Dansgaard-Oeschger cycles. One marks cool events and one warm.
Apparantly one of Gerard Bonds researches, Peter deMenocal has carried out further work to by looking at beryillium-10 and carbon-14 isotopes and apparantly the data tallies, debris from ice berg armadas tallies with changes cosmogenic isotopes in the ice cores. Hence most climate changes since the last ice age have been driven by this solar pulse amplified through feedbacks such as ice formation and changes in the intensity iof the ocean conveyor.
Re: #67
Because in statistics, that’s *always* the null hypothesis!
The reason for that is, that if was assume the data are a purely random process (no “signal” of any kind), then we can compute the probability distribution for whatever test statistic we apply. Then we can compare the measured test statistic to its probability distribution, to see how unlikely it is that the result is by random chance. If the result is too unlikely to be plausible (usually this means 5% probability or lower), then we regard the test as having contradicted the null hypothesis.
I quite agree. In fact I’ve tried to emphasize that just because the solar cycle doesn’t show in the temperature time series with statistical significance, doesn’t mean there’s no response. It just means that the response isn’t big enough to cross the detection threshold — yet.
As I said in an earlier post, I’d be *quite surprised* if the global temperature doesn’t respond to variations in TSI such as the solar cycle — as you point out, it’s basic physics. I just don’t think that solar variation is a plausible explanation for the observed large-scale modern global warming.
Re #30 (comment)
Gavin, I have no idea if there is any sort of link between the earth’s magnetic field and climate, but there is one interesting point on this: the earth’s magnetic field didn’t switch (stayed at “normal”, which is current polarity) during the Cretaceous, thus coincidences with a long era of warm climate…
[Response: People have looked for correlations between polarity reversals and climate and have found none. The Laschamp excursion (~40 kyr BP) was a near reversal, and there was no effect. Each of these events lead to big spikes in cosmogenic isotopes and hence incoming GCR. If GCR affects climate through cloud effects, where is the signal? The simplest conclusion is that there wasn’t one. – gavin]
Re: #60 YD & Heinrich events are strongest in the North Atlantic and are less evident in the Southern Hemisphere, in reply to #47, discussion of Rapid Climatic Change Events (YD & Heinrich, etc:).
Re: #68 Gerard Bond and evidence for a solar forcing event triggering, Little Ice age, Medieval warm period, Heinrich events, and YD.
Bond first. Gerard Bond et al’s Dec. 7, 2001 paper published in Science, “Persistent Solar Influence on North Atlantic climate, During the Holocene”
Comments:
1) Bond’s research was based on a study of deep-sea cores which show evidence of rapid periodic shifts in temperature at 200yr, 500yr and 1500yr intervals throughout the Holocene interglacial period, including the YD. Bond’s research found evidence of ice sheet discharges in the northern and southern Laurentide Ice Sheet and in Iceland all at nearly the same times, which challenged the hypothesis that Laurentide ice sheet changes, were responsible for the ice discharges. (Seager et al’s paper (see 47) challenges the fundamental assumptions of the thermohaline conveyor hypothesis explanation for the rapid climatic change events for other reasons. It has previously believed that periodic changes in the thermohaline conveyor caused a feedback loop on the Laurentide ice sheet which resulted in the ice discharges.)
2) Bond et al’s finding of evidence of periodic climatic changes in the deep sea cores, is supported by Paul Mayewski et al’s findings of evidence (increase in dust and sea salt in the ice core record, plus a dramatic drop in temperature (5C to 20C) on the ice sheet based on ratio of O16/O18) of rapid periodic climatic change events in the Greenland Ice Core, GISP2, also at similar intervals throughout the interglacial and Wisconsin glacial period. Interestingly Mayewski et al’s interpretation of the Greenland Ice core GISP is that the last interglacial period Eemian, ended abruptly.
From Bond’s paper “Our correlations are evidence, therefore, that over the last 12kyrs virtually every centennial time scale increase in drift ice documented in our North Atlantic records was tied to a distinct interval of variable, and overall reduced solar output (note Bond’s referenced change could be a solar electromagnetic type change, not a change in solar output, Astley).” “The climate change that accompanied the YD has not restricted to Greenland. The record of variations in the CH4 concentration (lower)… shows that tropical and subtropical climates were colder and drier during the YD.
Re: #60 Bond’s and Mayewski’s observed periodic climate variance is possibly due to two different but related periodic solar changes which directly and indirectly affect cloud formation in the upper atmosphere. The high latitude Northern and Southern upper atmosphere, is not symmetrical. The upper air temperature over Antarctic is much colder and drier. It is assumed the solar triggered upper atmosphere cloud formation is temperature dependent. If that were the case, there would be a greater effect in the Nothern hemisphere.
This seems to be a holistic problem (how can the sun change, how does different solar changes affect the atmosphere, how does the atmospheric changes, affect climate), that crosses disciplines. The climate record shows clear evidence of periodic rapid climatic cooling events. There are fundamental solar phenomena that are currently not explained. There is recent evidence of solar abnormalities. The problem with solar observation of 30 yrs and cycles of 200, 500, and 1500 yrs, is it is not possible to predicted what will happen based on what has happened in the last 30 yrs.
Re 65,66
Charlie, Grant, you didn’t mention the time period for which you calculated the spectrum for the CRU temperature data. Is it the same as for TSI (i.e. since 1978)?
If you just look at the spectrum, you should keep in mind that there are other important factors influencing global temperature on a similar time scale, mainly ENSO and volcanic eruptions.If you eliminate the ENSO signal from global temperature, you’ll find a different frequency spectum. The same for volcanic eruptions. For the time span 1978-2005 you’ll find a strong 9-year signal just from El Chichon (1982) and Pinatubo (1991). ENSO also contributes to the spectrum in the interannual time-scale. The superposition of volcanoes and ENSO might well produce an artificial 9-year signal (or 20 or whatever).
Moreover, especially if it’s over a longer time period (e.g. a century) the 9-year signal must have another source than the solar signal. I can’t see any reason why an 11-year solar cycle should produce a 9-year cycle in the temperature. There would be two more temperature cycles in a century than solar cycles, which is very hard to explain. Besides, that’s what you see in the Scafetta and West analysis….
This doesn’t mean that there is no solar signal. As mentioned before, you find some 11-year signal in the temperature cycle if you compensate for the ENSO and volcanic signal.
Maybe you could have a look at the spectrum of the volcanic and the ENSO signals.
Re 72,Urs, I looked at the full length of the series (Lean’s TSI from 1610 and the the CRU temps from 1850).
-Looking at White et al.(1977) it seems that solar signals arent very pronounced in the global average temp datsets, but can be found in the equatorial sea surface temperatures.-It would be interesting to know if Grant can find a truly signifcant signal in the GISST data.
Sadly I know very little about spectral analysis and so I cant comment on the 9 vs 11 year question.
Out of interest; Scafetta and West (2006) looks at the relationship between the Moberg reconstruction and Lean’s TSI reconstructions and suggests that 50% of the warming since 1900 could be down to the sun.
White,W. B., et al. (1997), A response of global upper ocean temperature to changing solar irradiance, J. Geophys. Res., 102, 3255â?? 3266.
Scafetta and West (2006), Phenomenological solar signature in 400 years of reconstructed Northern Hemisphere temperature record, GEOPHYSICAL RESEARCH LETTERS, VOL. 33, L17718, doi:10.1029/2006GL027142, 2006
Re 73
Charlie, we have to distinguish what we are discussing. Nobody dismisses an influence of the sun on earth’s climate, however, the question is where, when and how much. There will be regional differences in the influence and the corresponding correlation, since there are different processes involved in regional temperature variations.
Let’s stick to the influence on the global scale.
It is not a question of spectral analysis to see difficulties in linking two oscillating signals if one of the signals has two more cycles than the other within the same period. This you can’t explain by time-lag or phase shift or something alike. You could only explain that by the superposition of another cyclic signal which produces a shorter frequence. I haven’t seen any attempt on that. I don’t know if its possible to add another solar frequency to the 11-year cycle and end up with a 9-year cycle. I doubt it, but that’s indeed something for the spectral specialists…
Concerning the new Scafetta and West paper: It has similar problems as their other papers (see my comment 52 and link in there). They make some assumptions and pick out the data set which gives them the highest solar influence in the 20th century:
– They presume that all the temperature trend from 1600-1900 is due to solar forcing. Volcanic forcing is completely neglected. It’s not surprising, that their results show that solar forcing explains well the secular trend from 1600-1900… Besides they find a good match of multidecadal variations, which has been known before and was not really their topic.
– They choose the temperature reconstruction (Moberg et al.) which shows the highest long-term trend from 1600-1900 of all the reconstructions. This gives them the highest possible solar contribution in the 20th century. They only discuss the influence of choosing different TSI reconstructions but not the influence on choosing different temperature reconstructions.
In this paper Scafetta and West reveal an obviously sceptical agenda. They discuss an even higher solar influence listing well-known and antiquated sceptic’s arguments as the overestimation of 20th century warming due to heat-island effects, or the lower trends in satellite observations. And at last theyâ��ve found a new one: they suggest that the difference in the temperature increase over land and the oceans during the last decades might be due to contaminations of the land temperature recordâ�¦ – Well, this â��anomalous behaviourâ�� as they call it corresponds fully to what is physically expected. Maybe it would help to have a look at the IPCC report.
Re: #73, 74
Urs is quite right, that there’s no reconciling a 9yr period and an 11yr period in such a long time series without jumping through some major hoops. The 9yr period doesn’t quite make the cutoff for statistical signficance, so my guess is that it’s a random fluctuation.
I haven’t looked at the equatorial SST time series, but I’ll hit GISS and download it.
I see many are trying to find a 11 yrs signal in temperature series. Ok, T is perhaps the most important climatic parameter, but not the only one. Did someone ever look for a 11 yrs signal in long cloud cover time series? There are not much of them, perhaps, but I suppose astronomical observatories keep some record of cloud cover, isn’t it?
I also wonder if a high-frequency signal as that of the sunspot cycle could be really registered by SSTs: doesn’t the thermal inertia of the ocean dump it? I would be grateful to any oceanographer who could give me a precise answer to this question.
I am a casual reader of this site and was wondering what your thoughts were regarding this newly release data.
” Ilya Usoskin (Sodankyla Geophysical Observatory, University of Oulu, Finland) and his colleagues have investigated the solar activity over the past centuries. Their study is to be published this week in Astronomy & Astrophysics Letters. They compare the amount of Titanium 44 in nineteen meteorites that have fallen to the Earth over the past 240 years. Their work confirms that the solar activity has increased strongly during the 20th century. They also find that the Sun has been particularly active in the past few decades. ”
http://www.aanda.org/index.php?option=com_content&task=view&id=181&Itemid=42&lang=en
The following statement is from IPCC 2001:
Several recent reconstructions estimate that variations in solar irradiance give rise to a forcing at the Earth’s surface of about 0.6 to 0.7 Wm-2 since the Maunder Minimum and about half this over the 20th century… This is larger than the 0.2 W/m2 modulation of the 11-year solar cycle measured from satellites… All reconstructions indicate that the direct effect of variations in solar forcing over the 20th century was about 20 to 25% of the change in forcing due to increases in the well-mixed greenhouse gases.
Is there any change based on more recent research to this viewpoint? Am I correct to conclude the following:
The Medieval Warm Period and Little Ice Age were both mainly caused by changes in solar forcing, with some contribution from volcanic activity, and a good part of the warming in the first part of the 20th century was also solar driven. Most of the warming in the second part of the century was driven by the 2.4 W/m2 forcing from the increase in well-mixed greenhouse gases in the atmosphere.
Re 77
Thanks for the link. The paper does not present much new evidence. They present a test for the existing solar activity reconstructions which differ by a scaling factor, assuming that their values are the correct ones, because they are not influenced by terrestrial processes. However, they have to use model calculations based on chamber measurements to derive their relation between the Ti44 activity and solar activity. There seems to exist only one such model.
I’m not familiar with interstellar processes that could influence their data and do not know how accurate the model is. So the accuracy of their test leaves some questions.
Moreover, it seems quite difficult to improve the existing information about the trend over the 20th century, since the scatter between the few meteorites at the beginning of the 20th century is larger than the observed trend.
Let’s wait for comments by Muscheler et al., since their reconstruction has been ruled out by the test…
This press release should be relevant :
Getting closer to the cosmic connection to climate
A team at the Danish National Space Center has discovered how cosmic rays from exploding stars can help to make clouds in the atmosphere. The results support the theory that cosmic rays influence Earth’s climate.
http://spacecenter.dk/cgi-bin/nyheder-m-m.cgi?id=1159917791|cgifunction=form
I know that I’m entering an ethereal group of commentors that know their way around the theories of solar and atmospheric warming. But I’ll add my 3 cents anyway by asking this, “what studies have been made on just the level of heat pollution that we have been dumping into the system?”
I ask because of the direct BTU’s. Is there a study about the direct heat which we dump into the worldwide system? It seems to me that if it is a substantial amount, that can be added to the indirect means, (ie the other theories) then maybe we haave something.
Heat Fever: I did a back of the envelope calculation once which led me to believe that the direct impact of direct heat was about 1/40th of a W/m^2, which is about two orders of magnitude less than that contributed by anthropogenic GHG forcing.
You could probably repeat my calculation by looking up worldwide BTU energy consumption/production and dividing by the area of the earth.