Guest post by Alec Rawls
When he argues that a reduction in forcing will cause cooling Dana Nuccitelli is not actually talking about drinking. He is talking about the solar forcing of global temperature, but the drinking analogy is a handy way to understand where his argument goes off the rails.
Mr. Nuccitelli, who blogs for the consensus-approved Skeptical Science website, was writing about Henrik Svensmark’s GCR-cloud theory of indirect solar forcing, where Galactic Cosmic Rays are hypothesized to seed cloud formation. If Svensmark is right then a strong solar wind, by deflecting some GCR from reaching the earth, in-effect blows some of the clouds away, letting more sunlight through to warm the planet’s surface.
That can’t possibly explain late 20th century warming says Nuccitelli, because GCR deflection has been estimated (see the Krivova-Solanki graph above) to have peaked by 1980. The raw data suggests the actual GCR minimum was ten years later, but set that aside. Nuccitelli thinks it is the change in the level of forcing, not the level of forcing, that determines whether the climate system warms or cools:
So, if GCRs really do amplify the solar influence on global temperatures, since 1980 they are amplifying a cooling effect.
Cooling begins when a forcing passes its peak? Fail. Daily temperatures don’t start falling at noon. They continue rising until mid-afternoon. The hottest time of the year isn’t the first day of summer (the summer solstice, after which the days start getting shorter), the hottest time is mid-summer. To think cooling should start when forcing passes its peak is like thinking you can sober up by drinking just a little more slowly.
Here’s a tip for Dana to keep in mind on New Year’s eve: it is the level of alcohol forcing that matters. If you are drinking alcohol faster than you body is excreting it (not exactly the way the earth excretes heat, but similar enough), then your blood alcohol is rising. You are getting drunker, even if you have lowered the rate of your drinking! That’s right, putting a little less rum in your egg nog will not sober you up! Your increasing inebriation will just be a little less rapid, and it is the same for solar forcing.
When the peak level of forcing appears in the rearview mirror, the downward trend in the forcing that begins at that point does not cause cooling. It just causes warming to be a little less rapid. Only when the energy pouring into the climate system falls to the level of the energy escaping back out does the system stop warming. Empirically, that turns out to be mid-afternoon, mid-summer, and approximately the first decade of the 21st century.
Three blind mice
Dana Nuccitelli produced one of three widely cited rebuttals to my suggestion that a new sentence that was added to the Second Order Draft of AR5, a sentence that admits strong evidence for some substantial mechanism of solar amplification, is a “game changer.” That admission is on page 7-43 of the SOD:
Many empirical relationships have been reported between GCR or cosmogenic isotope archives and some aspects of the climate system … The forcing from changes in total solar irradiance alone does not seem to account for these observations, implying the existence of an amplifying mechanism such as the hypothesized GCR-cloud link.
That’s a game changer because the only solar forcing included in the IPCC computer models is the very slight variance in solar irradiance (also known as TSI, or “the solar constant”). If there are other solar forcings in play, working through variables that actually vary substantially as solar activity ramps up and down, that kills the report’s key finding (on page 8-4) that we can have “very high confidence that natural forcing is a small fraction of the anthropogenic forcing.”
The two most widely cited rebuttals, which I answered last week, were both by lead authors from the IPCC. Steven Sherwood, one of 15 lead authors of chapter 7, pretended that the admission of evidence for “an amplifying mechanism” was only about GCR-cloud. He then proceeded to claim that the evidence for GCR-cloud points to a weak mechanism, and used that as a grounds for dismissing the idea that any substantial solar forcing beyond TSI could be at work.
Doesn’t follow. The evidence for “an amplifying mechanism” (emphasis added) is entirely separate from the evidence for the GCR-cloud mechanism. The former is paleo evidence, where numerous studies of the geologic record have found strong correlations between solar activity and climate going back many thousands of years. The evidence for the GCR-cloud mechanism is from cloud-chamber experiments and ongoing observations of cloud micro-physics.
It doesn’t matter how unconvinced Sherwood is by the evidence for the GCR-cloud mechanism. That evidence does nothing to counter the paleo evidence, cited in the draft report, that some mechanism of enhanced solar forcing must be at work. By using his discontent with the GCR-cloud theory as an excuse to dismiss the paleo evidence, Sherwood is inverting the scientific method, and he is lying to the public about what the report says, making him a seriously bad guy.
Apparently weak minds think alike because Nuccitelli did the same thing Sherwood did, only a day earlier. Dana’s post only looks at the GCR-cloud mechanism and completely ignores the draft report’s admission of strong evidence for some mechanism of solar amplification. It is in the context of that more fundamental mistake that Nuccitelli goes on to completely misinterpret the evidence for the GCR-cloud mechanism itself, claiming that anything less than peak forcing causes cooling, arguing in-effect that he can sober up by drinking a little slower. Just tell that to the officer Dana. He won’t even need to give you a breathalyzer.
More evidence that weak minds think alike is the second semi-official rebuttal to my “game change” claim, issued by Joanna Haigh, a lead author of the IPCC’s third report. Haigh proceeds on the same dishonest pretence as Sherwood, telling NewScientist magazine that the new sentence in the draft report is only about GCR-cloud, which she then dismisses with the same drinking-game mistake that Dana makes, claiming that if climate were being driven by solar activity then the planet would have started cooling when solar activity was at its peak:
Haigh points out that the sun actually began dimming slightly in the mid-1980s, if we take an average over its 11-year cycle, so fewer GCRs should have been deflected from Earth and more Earth-cooling clouds should have formed. “If there were some way cosmic rays could be causing global climate change, it should have started getting colder after 1985.”
Sober up Joanna. Have a single shot instead of a double. Works every time.
Which theory has more trouble with flat 21st century temperatures?
It is amusing how Dana Nuccitelli, through sheer incompetence, was able to prefigure the highly credentialed malfeasance of both of these IPCC fraudsters. Still, Nuccitelli has to be credited with at least a bit of misfeasance of his own because he wasn’t satisfied with just assuring his gullible readers that cooling commences when forcing is at its peak. That only supplied an excuse for dismissing a solar explanation for late 20th century warming, leaving the conspicuous lack of 21st century warming still to be dealt with. Dana’s solution? Pretend that the flat 21st century temperature record militates against a solar driver of climate:
In fact, GCRs reaching Earth recently hit record high levels (Figure 4), yet temperatures are still way up.
Temperatures have merely flattened out, they haven’t gone down yet, and no Skeptical Science reader will ever learn that this is just what the discovered correlations between solar activity and climate predict. The strongest temperature response to a change in solar forcing is seen with a lag of about ten years (Usoskin et al. 2005), or one solar cycle (Solheim et al. 2012). The theory that is discomfited by flat 21st century temperatures is the CO2-warming theory, which predicts ever more rapidly increasing temperatures.
Dana might actually think that the flat 21st century temperature record causes trouble for the solar-warming theory but there is no way he can think it causes less trouble for the CO2-warming theory. For him to pretend that 21st century temperatures favor the CO2-warming theory is inexcusably dishonest, but as usual, the professionals are even worse. Note this little gem from the SOD (p. 7-44):
The lack of trend in the cosmic ray intensity over the last 50 years (Agee et al., 2012; McCracken and Beer, 2007) provides another strong argument against the hypothesis of a major contribution of cosmic rays to ongoing climate change.
That’s 15 IPCC lead authors all accepting the crackpot idea that you can only get drunk if your rate of drinking is going up. Steady exposure to the high 1950’s level of solar activity will keep you from warming, just as steady swilling of a high level of booze will keep you stone-cold sober. But where Dana only said that “temperatures are still way up” (implicitly acknowledging that they are no longer going up), the draft report here claims that temperatures are still going up (“ongoing climate change”).
Must be a Steven Sherwood sentence. He needs to look at page 10-3 of the SOD (emphasis added):
While the trend in global mean temperature since 1998 is not significantly different from zero, it is also consistent with natural variability superposed on the long-term anthropogenic warming trends projected by climate models.
Note that the “consistent with natural variability” part is a near call, after NOAA admitted in 2008 that 15 years of no warming would falsify current models. But yeah, let’s pretend it is the solar theory that has trouble with the lack of recent warming.
The Guardian, Andrew Sullivan, DeSmog, Romm etcetera, all pwned by Dana Nuccitelli’s error-filled AR5 post
The ensuing Skeptical Science newsletter bragged about how many eco-propagandists picked up on nutty Nuccitelli’s non-stop nonsense and the list is indeed impressive, a glaring testament to the total absence of due diligence on the part of these “journalists,” none of whom thought to question Dana’s advice on how to sober up quick. Here is the SkS tally of eco-scalps:
This was a very big week for SkS in the news. Dana’s IPCC Draft Report Leaked, Shows Global Warming is NOT Due to the Sun was re-posted and/or linked to by The Guardian, New York Times Green, New York Times Dot Earth, Huffington Post, Climate Progress, Mother Jones, Climate Crocks, Carbon Brief, Grist, Daily Beast, DeSmogBlog, Graham Readfearn, Der Spiegel, Maribo, Learn from Nature, Alternative Energy in the 21st Century, and Motherboard. It was also Tweeted by Michael Mann and Chris Mooney, among many others.
The only “consensus” journalist on this list who showed any integrity was Andrew Revkin, who had already written a post on my leak of the draft report. After updating that post with a link to Nuccitelli, Revkin updated again later with a link to the rebuttal that Jo Nova and I wrote about Seven Sherwood. Thanks Andrew, for being an actual journalist.
To put their “big week” in perspective the folks at Credulous Science reached high for sufficient words:
Winston Churchill once said: “A lie gets halfway around the world before the truth has a chance to get its pants on.” Not this time; we got the truth’s pants on in record time and nipped this myth in the bud before the contrarians were able to misinform the public.
The “truth” in their rendering is whatever patent falsehoods can be used to fool the ignorant into thinking they should fear CO2. No, nutty ones. You did not forge an exception to Churchill’s dictum. You provided a textbook example of it.
Dana is a poster child for those who feel a moral imperative to “believe the scientists”
For the last two years I have had a lot of fun exposing the large number of top climate scientists who claim that it is not the level of forcing that causes warming, but the trend in the forcing. They are all looking at the wrong derivative (one instead of zero).
Given the enormous pressure on the eco-left to accept what these government-funded climatologists are saying it is not surprising that someone like Dana Nuccitelli would swallow the idea that temperature really is driven by the trend in the forcing, and one super-wacky segment in Dana’s post indicates that he really is accepting that this is how physics works. If you leave a pot of water on a steady flame it won’t heat. If you want to heat the water you have to keep turning the flame up. The segment is titled, “Physical Reality Intrudes on Rawls”:
Rawls has argued to the contrary by claiming that the climate is still responding to the increase in solar activity from the early 20th century, and that GCRs are amplifying that solar warming from over 60 years ago. This argument is simply physically wrong. As Figure 2 illustrates, when solar activity rises, temperatures follow suit very soon thereafter. In fact, during the mid-20th century, solar activity and global surface temperatures both flattened out. Are we to believe that the planet suddenly began responding to the pre-1950 solar activity increase in 1975—2012, after not warming 1940—1975? The argument makes no physical sense.
Obviously I never said that late 20th century warming was caused by solar activity from the early 20th century and Dana does not give a citation for what argument of mine he is referring to but its easy to figure out. I have argued many times that if one combines the strong paleo correlations between solar activity and climate with the fact that solar activity was at what Ilya Usoskin calls “grand maximum” levels from 1920 to 2000 then it is certainly plausible that much of 20th century warming, including late 20th century warming, could have been caused by the sun. 80 years of a high level of enhanced solar forcing just might warm the place up a bit (and it only did warm a bit, about 0.8 °C over the century).
If I am attributing late 20th century warming to the high solar activity that persisted through 2000, why does Dana think I am attributing it to solar forcings from 60 years earlier? He must be fixed on the idea that only a change in the level of solar forcing can cause warming. That’s what all of these top scientists have been telling everyone and there was no rise in solar activity after 1950, hence any solar-caused warming would have to stem from the pre-1950 rise in solar activity.
But come on Dana. That is not what I am saying. That is what they are saying, and I have been trying my darnedest to expose it as a blatant misrepresentation. I’m not attributing late 20th century warming to the pre-1950 rise in solar activity. I’m attributing it to the fact that solar activity remained at close to the same high 1950’s level until 2000 (or 2003). It is the level of the forcing that causes warming, not the trend in the forcing.
I have to feel bad for Dana on this point. It isn’t his fault. He has been systematically duped by this parade of so-called scientists all telling him that a persistent high level of forcing can’t cause continued warming. Makes me want to put him on a milk carton. The poor guy isn’t just lost, he was kidnapped. Want a piece of candy little boy? Credulous Science indeed.
lgl says:
January 4, 2013 at 3:08 pm
The temperature to go down the next few decades if TSI and solar activity goes down is exactly what to expect
So temperature varies as TSI…
lgl says:
January 4, 2013 at 3:08 pm
How likely such a fit is to be generated by chance?
Very unlikely, http://virakkraft.com/TSI-integral-Loehle-temp.png
It is not by chance at all. After you have removed [by smoothing over a 200-yr window] all known solar cycles 11, 22, 88, 120, 200 yr, what is left is the 1000-2000 climate cycles. So you are comparing climate with climate. There is no evidence that these cycles are solar, your plot is good evidence they are climate.
Lief I know you have strong opinions on this, but FWIW I believe it can be demonstrated that the ocean response time is sufficiently long to prevent a full response to the energy imbalance imparted with each solar cycle. At the onset of the next cycle, the residual imbalance remaining from the previous cycle carries over to the next causing a cumulative effect. This explains why integration of tsi or solar proxies can generate interesting correlations with non stationary temperature series even when the underlying function (the solar cycle) has little if any trend.
The effect can be demonstrated with simple white noise. Here is a white noise series (n=10,000 mean=0, trend=0). Here is the same series integrated. I don’t believe that OHC should be modeled as a pure random walk such as the white noise example because such series are unbounded and therefore unphysical, however the autocorrelation of OHC suggests that it’s memory period extends far beyond the typical 11 year cycle length and is consistent with an integrated input.
Here is the integral of normalized Moscow cosmic ray data plotted alongside a time series of ocean temperatures at 30 meter depth. Before integrating the cosmic ray data I offset the mean (0.4 sigma if I recall) which improves the fit somewhat presumably because it centers the series near the energy balance point.
Layman Lurker says:
January 4, 2013 at 5:42 pm
Lief I know you have strong opinions on this, but FWIW I believe it can be demonstrated that the ocean response time is sufficiently long to prevent a full response to the energy imbalance imparted with each solar cycle.
I don’t really disagree with that. The way I would put it is that every frequency of the driver creates its own signal, so there will be an 11-yr signal in temperature [people actually claim to have found one], and a 120-yr signal in temperature and so on. If you integrate [or just smooth over a window – same thing] you suppress the high-frequency stuff. All that is basic stuff. When you as lgl smooth over 200 years you suppress all known solar variations. What is left is the longer [e.g. 1000-2000 year variations of the climate system itself, so no wonder that correlates with the long-term temperature series. There is no evidence that the 1000-2000 yr ‘cycles’ are of solar origin, and, as lgl shows, good evidence that they are of climate.
Perhaps I’m misunderstanding you? Integrating is definitely not the same thing as smoothing. Nor would an 11 year MA smooth do the “same thing” as integrating the solar cycle (over whatever period you like). If oceans integrate the solar signal, then we should not expect ocean temperature data to look like the underlying solar function. If oceans integrate the solar signal then the proper regression model specification would be to compare the integral of solar (or cosmic ray etc) to OHC. Note – I’m not suggesting that the Steinhilber series does not suffer from climate contamination.
Leif
After you have removed [by smoothing over a 200-yr window] all known solar cycles 11, 22, 88, 120, 200 yr
I have not. http://virakkraft.com/TSI-integral-Loehle-temp.png
your plot is good evidence they are climate
It is the TSI integral that correlates with temperature, not the TSI, and the integral lags the signal by around 1/4 signal period, several decades for the 200 yrs cycle. Climate can’t affect something that happened in the past, not even in your universe.
Layman Lurker says:
January 4, 2013 at 10:14 pm
Perhaps I’m misunderstanding you? Integrating is definitely not the same thing as smoothing.
When I asked what they meant by ‘integration’ the answer was ‘a running sum’ and it seemed that the ‘running sum’ should be taken over a window of 200 years. My interpretation of that is that ones should sum the values [reduced by the overall mean] over the 200 years, and assign that value to the last year. Then move back one year and repeat, etc. Since the interval has constant length [200 years], the sum is proportional to the average, i.e. the procedure is equivalent to a simple moving average. The moving average has the advantage of being independent of the resolution of the data. This is what I call smoothing.
lgl says:
January 5, 2013 at 4:36 am
Climate can’t affect something that happened in the past, not even in your universe.
Past climate can.
Leif
It isn’t past climate. TSI ’causes’ the TSI integral, A causes B. B correlates with C, temperature.
Then your absurd conclusion is: B correlate with C which shows C is causing A.
About your smoothing procedure. Use these ‘solar cycles’: http://www.woodfortrees.org/plot/hadcrut4gl/mean:36/derivative/mean:36
and try your procedure. And try some smoothing:
http://www.woodfortrees.org/plot/hadcrut4gl/mean:36/derivative/mean:36/mean:360
Did that give you the correct temperature record?
Then add some very minor average to the first and see what happens to the integral.
lsvalgaard says:
January 5, 2013 at 4:40 am
Thanks for the explanation Leif. I believe I understand where you are coming from now. The problem I have with this integration window approach is arbitrary centering by straight up 200 year window mean subtraction. In physical terms, the series should center where energy balances. Obviously, this does not necessarily equate to the 200 year mean (or whatever the proper window is) and would seriously misspecify the actual relationship to climate. Also, phenomenon like feedbacks, volcanoes,etc would play into this as well.
lgl says:
January 5, 2013 at 11:26 am
Then your absurd conclusion is: B correlate with C which shows C is causing A.
My conclusion is that when you remove all known solar cycles 11, 22, 88, 120, and 200 years you are left with the longer 1000-2000 year climate cycles.
Now, if you don’t believe that then let’s take another approach and compare with the temperature in central Greenland where the ice cores come from:
http://www.leif.org/research/Greenland-Temp-Steinhilber-200yr-png
Perhaps that data is also no good where the correlation breaks down?
Layman Lurker says:
January 5, 2013 at 11:26 am
Thanks for the explanation Leif. I believe I understand where you are coming from now. The problem I have with this integration window approach
I have problems too. I have tried to get lgl to explain what he is doing and what I explained is what I got out of his vague description. Perhaps he could be more precise.
lsvalgaard says:
January 5, 2013 at 11:41 am
http://www.leif.org/research/Greenland-Temp-Steinhilber-200yr.png
Leif
Perhaps that data is also no good
Right, the temp of Greenland is determined by Dansgaard-Oeschger events. How does it correlate with Loehle by the way?
I’m doing this like you explained. Running sum of the last 200 yrs, but without removing any average, you remove the amplitude then, the amplification(?). (Peak-to-peak of the integral is around 80). And a simple smoothing does not give any lag so it will not match the temp record.
Didn’t you see this in the woodfortrees examples?
lgl says:
January 5, 2013 at 1:51 pm
Right, the temp of Greenland is determined by Dansgaard-Oeschger events. How does it correlate with Loehle by the way?
Perhaps the other way around. The Greenland data matches Loehle and you the last 1000 years.
I’m doing this like you explained. Running sum of the last 200 yrs, but without removing any average
You said something about subtracting 5. What was that about?
A running sum is the same as smoothing with the result assigned to the last year.
lgl says:
January 5, 2013 at 1:51 pm
Right, the temp of Greenland is determined by Dansgaard-Oeschger events. How does it correlate with Loehle by the way?
………………..
If you compare change in the Earth’s magnetic field (across 3 or 4 decades) reflecting the tectonic movements just off Greenland’s coast in the Kolbeinsey ridge, you will find pretty good mach to the Loehle’s temperature reconstruction.
http://www.vukcevic.talktalk.net/LL.htm
Of course Dr. S. will tell you it is just appearance, we have no data, pseudoscience or whatever else he may think of.
Good night.
Leif
Between 0 and 1000 you mean? Not very good after 1000.
The subtracting 5 was that other methode, yours is better, I didn’t worry about the amplitude then, only the shape. Smoothing is not “with the result assigned to the last year”. It must be centered around the mid year.
Leif writes “The difference in the resulting temperature is about a tenth of a degree. That is minute.
Look carefully at the label on the axes of the left side panels. What does it say?”
I’m less interested in what a model says the temperature difference will be and more interested in the pure W/m2 value and that appears to be the order of tenths of W/m2 (/um) from those graphs.
So simply put, is it in fact the order of several tenths of W/m2 or not Leif?
And do you recognise that the Ocean warming forcing is about 0.6 W/m2 or do you think there is some other value that should be used?
TimTheToolMan says:
January 5, 2013 at 7:07 pm
that appears to be the order of tenths of W/m2 (/um) from those graphs.
It says W/m2/nm that is W per m2 per nanometer
Leif writes “It says W/m2/nm that is W per m2 per nanometer”
You’re right, it does say per nm rather than per um. But you’ve not answered the question because per um or per nm is only a distraction as to the actual difference in W/m2 the new method makes. So I’ll just ask you straight out as they’re your graphs.
How many W/m2 difference is there between the two methods?
TimTheToolMan says:
January 5, 2013 at 10:40 pm
Leif writes “It says W/m2/nm that is W per m2 per nanometer”
You’re right, it does say per nm rather than per um.
Either way, I commented because you were all to happy to get some confirmation bias for your 0.6 W/m2. To answer the question you have to integrate [you know: integrate] over the wave length region for UV, VIS, and NIR used in the Figure. I don’t know, offhand, what they were. The point is that the people who wrote the paper [from which I got the graphs] did this and calculated how much difference it would make in temperature. As you can see: about 0.1 degree, so the Harder Effect is minute, even if real.
Leif writes “Either way, I commented because you were all to happy to get some confirmation bias for your 0.6 W/m2. ”
I dont have a particular need to see 0.6 W/m2. Levitus found is was 0.39 W/m2. Basically its a low number of W/m2 that is heating the ocean and if the new SIM data reveals variations in the tenths of W/m2 range, then I want to know about it to further my understanding of what could be happening. I want to be unbiased in that overall view.
Frankly its too early to know for sure anyway. We simply dont have enough SIM data and any “corrections” (eg Harder) are little more than guesses. I understand that too.
I can see from the graph that the earlier assumption was that (over the same per nm ranges) the variations appear to be very small. Perhaps 0.1W/m2. And I can see that after the Harder adjustments over the same per nm ranges the variations are now several tenths of W/m2.
I am uninterested in what people suggest is the resulting temperature for a given x W/m2 increase because frankly its a silly measure. The only real indicator of global warming is the earth’s (ie ocean’s) accumulation of energy and that’s the figure I try to look for.
Tim the toolman says
The only real indicator of global warming is the earth’s (ie ocean’s) accumulation of energy and that’s the figure I try to look for.
Henry@timthe toolman
the sun heats the oceans mostly with radiation of the UV type, because water (spectrum) shows strong absorption in the UV region.
Once you understand that right you can follow easily what I say here
http://blogs.24.com/henryp/2012/10/02/best-sine-wave-fit-for-the-drop-in-global-maximum-temperatures/#comment-192
more clouds and rain in a certain latitude band is a natural consequence of more (natural) cooling (as evident from the flooding Nile)
once you understand we are in a cooling period
which we are
http://www.woodfortrees.org/plot/hadcrut4gl/from:2002/to:2013/plot/hadcrut4gl/from:2002/to:2013/trend/plot/hadcrut3vgl/from:2002/to:2013/plot/hadcrut3vgl/from:2002/to:2013/trend/plot/rss/from:2002/to:2013/plot/rss/from:2002/to:2013/trend/plot/gistemp/from:2002/to:2013/plot/gistemp/from:2002/to:2013/trend/plot/hadsst2gl/from:2002/to:2013/plot/hadsst2gl/from:2002/to:2013/trend
note that even hadsst2 is on its way down