Error Theory

Solar warming and ocean equilibrium, part 4
Crossposted at WUWT

I emailed Dr. Muscheler about the very strange remarks that were attributed to him in the recently released report on last year's NCAR workshop: The Effects of Solar Variability on Earth's Climate. Dr. Muscheler says that the report's version of his remarks "is obviously a mistake," but he answers my query about what he had meant to say with yet another obvious mistake, a mistake the greatest import, and one which is no less egregious for being widespread.

The report (available for free download from the National Academies Press) seems to paraphrase Dr. Muscheler as claiming that cosmic ray flux during the late 20th century was "steady and high" (p. 17):

Muscheler stated that proxy data indicate that the cosmic-ray flux actually decreased early in the 20th century, but recently the level has been steady and high. Based on the proposed link between increased GCR flux and cloudiness, one might have expected that the late 20th century would be cooler than the early 20th century—a state that was not observed.

So I asked Dr. Muscheler:

By this paraphrase, your comment that "recently the level has been steady and high" seems to be referring to "the late 20th century," but that can't be right.

The abstract that you provided for your remarks begins by describing cosmogenic radionuclides as "the most reliable proxies for reconstructing solar activity variations thousands of years back into the past" (p. 41). But late 20th century solar activity was high, so if GCR is actually a proxy (inverted), it must have been low in the late 20th century [or it isn't much of a proxy].

Usoskin 2007 estimated a grand maximum of solar activity from 1920 to 2000. Lockwood put the peak of this grand maximum in the mid 80's. Thus the statement attributed to you has to be a mis-transcription of some sort.

I'm guessing that your remark about recent GCR flux levels being "steady and high" was actually a reference to post 2003, not to "the late 20th century." But that leaves the question of on what grounds you were claiming that the late 20th century should have been cooler than the early 20th century, or did they mis-transcribe that as well?

... If you really do think that, according to the GCR data, the late 20th century should have been cooler than the early 20th century, can you please explain why?

Of course I know the highly unscientific grounds on which numerous "consensus" climate scientists make such claims, but it's important to get them on record saying it.


You can't heat a pot of water by turning the flame to maximum and leaving it there

It's not the level of the flame that causes warming, but the rate of change in the level of the flame. Everybody knows that, or so the anti-CO2 establishment would have us believe. See for instance, Rasmus Benestad, 2005:

A further comparison with the monthly sunspot number, cosmic galactic rays and 10.7 cm absolute radio flux since 1950 gives no indication of a systematic trend in the level of solar activity that can explain the most recent global warming.

It doesn't matter that solar activity was at grand maximum levels from 1920 to 2000. Only the continued turning up of a forcing can cause warming according to Dr. Benestad.

Here is a list of a dozen more top consensus climate scientists all making similar goofball statements, and as I discovered from my "expert review" of the First Order Draft of AR5, this is now the IPCC's official grounds for dismissing a solar explanation for late 20th century warming.

Would Muscheler add himself to the list? I had to give him a chance and he very graciously took it, thanking me for pointing to the obvious error in the transcription while confirming that, yes, he too looks at the wrong derivative. He should be looking at the zero derivative (the level of solar activity) but is instead looking at the first derivative (the rate of change in solar activity, or the trend).


Muscheler's response (emphasis added)

Dear Alec Rawls,

unfortunately I haven't been involved in writing this report. This statement is obviously a mistake and I don't know why it ended up in the report.

In the early 20th century solar activity increased and, therefore, the cosmic ray flux decreased. According to the cosmic ray-cloud hypothesis the (low) clouds should have decreased and it should have led to a warming.

Solar activity & cosmic rays were relatively constant (high solar activity, strong shielding and low cosmic rays) in the second part of the 20th century and, therefore, it is unlikely that solar activity (whatever process) was involved in causing the warming since 1970.

Maybe I was unclear in replying to a question or there was a misunderstanding from the person writing the report. Anyway it is obviously wrong in the report.

Thank you for making me aware of this problem. I will contact the authors and ask if it can be corrected.

Best wishes,

Raimund Muscheler

The hidden (and completely untenable) assumption of rapid ocean equilibration

Last year I emailed the dozen climate scientists from my list of those who have made these kinds of claims and suggested that they must be assuming that that by 1980 or so the oceans had already equilibrated to whatever temperature forcing effect high 20th century solar activity might be having, otherwise the continued high level of forcing would cause continued warming.

Several confirmed that they were indeed assuming rapid ocean equilibration to any change in climate forcing. One was Mike Lockwood, whose 2007 paper with Claus Fröhlich had opened with a strong assertion that it is the trend in a forcing, not the level of a forcing, that causes temperature change:

There is considerable evidence for solar influence on the Earth's pre-industrial climate and the Sun may well have been a factor in post-industrial climate change in the first half of the last century. Here we show that over the past 20 years, all the trends in the Sun that could have had an influence on the Earth's climate have been in the opposite direction to that required to explain the observed rise in global mean temperatures.

If the paper was assuming rapid ocean equilibration it really ought to have said so, but better late than never. In his response to me Lockwood offered evidence that ocean equilibration takes at most a decade, but his estimate does not stand up to scrutiny. It was derived from an energy balance model (Schwartz 2007) that represents the oceans by a single heat sink.

This is a highly unrealistic simplification (having the whole ocean change temperature at once). If a more realistic 2-heat-sink model is used, where it takes time for heat to transfer from one ocean layer to another (Kirk-Davidoff 2009), then the observed rapid temperature adjustment of the upper ocean layer tells us next to nothing about how long it takes for the ocean to equilibrate to a long term forcing. (Full discussion in Part 2 of my "solar warming and ocean equilibrium" series.)

The Lockwood and Fröhlich paper acknowledges that there was a long natural warming from the bottom of the Little Ice Age (punctuated by notable downturns when solar activity fell during the Dalton Minimum and around the beginning of the last century), and they say themselves that this long natural warming was probably caused by increasing solar activity, yet we are supposed to be confident, on the basis of a completely unrealistic one-heat-sink model, that this long warming just happened to end in 1980, when the whole idea of a long period of solar warming is fundamentally inconsistent with that model. Crazy.


Workshop participant Isaac Held: "equilibration takes centuries"

One of NCAR's workshop panelists actually addressed the time-to-equilibration issue (p. 21, emphasis added):

Issues in Climate Science Underlying Sun/Climate Research
Isaac M. Held, National Oceanic and Atmospheric Administration
Geophysical Fluid Dynamics Laboratory

In his presentation Isaac Held asserted that the response of the climate to radiative heating—whether it comes from greenhouse gases trapping heat, stratospheric aerosols from volcanic eruptions or aerosols of various origin reflecting sunlight back to space, or finally variable TSI heating—involves both the troposphere and the ocean. The surface and the troposphere are intimately coupled through fast radiative-convective adjustments so that they respond as a whole, with part of the heat input going into the ocean. The ocean heat uptake and later slow release back to the atmosphere are the factors responsible for the difference between the transient response of the climate to radiative forcing as compared to the equilibrium climate (some 40-70 percent of the adjustment is achieved on a timescale on the order of 4 years, whereas equilibration takes centuries). This transient behavior can be demonstrated using a simple two-box model of the mixed layer and deep ocean, and it applies to all radiative forcings, such as to the Mount Pinatubo volcanic aerosols, as well as for the response to the 11-year solar cycle. On stratosphere-troposphere coupling, there is recent observational evidence that in the Southern Hemisphere the surface westerlies (and the storm track) have shifted poleward by a few degrees due possibly to the ozone hole over the South Pole in the stratosphere.

[Stephen Wilde will want to look at what the report says about the rest of Held's presentation.]

To clarify, when there is a long term change in forcing it isn't 40-70 percent of the eventual deep ocean heat storage that is achieved within four years. Here Held is talking about the time-response of GMAST (the Global Mean Air Surface Temperature), which is largely driven by ocean surface temperature, and the ocean surface warms up quickly in response to an increase in forcing.

If elevated forcing persists for decades or centuries this warmed-up upper ocean layer will all-the-while be transferring heat to deeper ocean depths, causing the temperature differential between the upper and lower layers to shrink which in turn causes a slowing of the heat loss from the upper ocean to the deeper ocean. That slow decrease in heat loss from the upper ocean layer causes the upper ocean layer to slowly get warmer, which in turn causes a slow increase in atmospheric surface temperatures (the remaining 30-60 percent of the GMAST increase that Held is referring to). This continued warming can go on for centuries.

So I must appeal to Dr. Held: you really need to point out to your colleagues the implications of moving to a more realistic "two box model" (never mind a 3 or 4 box model) where it takes time for heat to accumulate in deeper ocean layers. If prolonged forcing can cause the oceans to warm for centuries (and GMAST to continue to rise for centuries) then no, we cannot be confident that by 1980 the oceans had equilibrated to the 20th century's grand maximum levels of solar activity.

This is regardless of whether those levels were pre or post peak. It's the level that matters, not the trend.


A helpful diagram

If anyone has trouble understanding why they should be looking at the level of a hypothesized solar-magnetic forcing, not just the trend, here is a helpful diagram from Ken Gregory:



Temperature falls only when the level of forcing falls below that needed to maintain the current temperature. With typical cyclical behavior, temperature peaks often lag considerably behind peaks in forcing. Everybody is familiar with this phenomenon from daytime temperatures, which do not peak at noon but peak in the mid-afternoon. So too with longer period forcings and deeper heat sinks.

So no, if temperature continues to rise after solar forcing has peaked it does not indicate that the continued warming is not caused by solar forcing. On the contrary, it is exactly what we would expect from a solar driver of climate.

In the case of late 20th century solar forcing there really was no discernable peak but rather a 50-year plateau, in which case temperatures should continue to rise until equilibrium is reached. There is no reason to think the oceans would have equilibrated to high 20th century forcing by 1980, and so no reason to dismiss a solar explanation for post 1980 warming.

Day vs. Season

In part 3 of my series Solanki and Scheussler offered a different rationale for assuming rapid ocean equilibration. The strong correlation between solar activity and climate that they had found was strongest with a short lag, less than ten years, so if there were longer-term solar effects, these scientists insisted that they had found no evidence for it. But that is wrong. Rapid responses to solar forcing are evidence for longer term responses, just as the rapid daytime temperature response to the rising sun implies that the hemispheres should warm when their seasons progress towards the greater insolation of summer.

This is pretty basic stuff so maybe these guys just aren't getting out enough. They don't talk to people who don't share their eagerness to grab at any rationale that supports the CO2-warming theory, no matter how patently weak it is. And its pretty clear they aren't even talking about these things amongst themselves.

Not a one of the quotes I have compiled betrays any hint of hidden assumptions about rapid ocean equilibration or anything else. They are unconditioned statements: the solar flame was not rising so it could not have caused warming. Only when pressed by WUWT do they scramble to support their unstated premises.

For each of these scientists it seems that plan-A was that nobody would notice that they were looking at the wrong derivative.


Leif Gets It right (right Leif?)

On the other side we have everyone who has ever heated a pot of water, including our own Dr. Leif Svalgaard, who was provoked last month to admit:

When I start the pot in the morning on maximum in order to get hot water for my tea and to boil my eggs, it works great for me. I get hot tea and boiled eggs in minimum time. If I turn down the heat, it takes longer…

Good thing Leif has tenure already. His mundane observation rebuts the very heart of the anti-CO2 industry's dismissal of solar-driven warming.

Another question for Muscheler

When 50 years of steady high solar activity coincide with rising average temperatures, that would seem to be evidence for a solar driver of climate. What is Raimund Muscheler's grounds for taking it as evidence against? His response to my first query does not say, so I sent him a second. I am contacting Isaac Held as well, whose 2 cents would be much appreciated.

Maybe Raimund really does think that it is the rate of change of a forcing rather than the level of a forcing that causes warming but I doubt it. More likely he has accepted the rapid-ocean-equilibrium assumption of Lockwood, Solanki and others without thinking it through. (Note that the paraphrase of Muscheler's comments in NAP's NCAR report has him making the same assertion as Lockwood: that if the sun were driving global temperature then late 20th century temperatures should have been falling, not rising. That seems to indicate a Lockwood-like rapid equilibrium assumption)

Muscheler's 2007 paper on paleo and recent GCR deposition suggests that 20th century solar activity was merely "high instead of exceptional," but for time-to-equilibration this distinction makes no qualitative difference. It is true that the smaller the change in forcing the faster equilibrium should be reached (like starting partway in on the equilibration response to a larger change in forcing), so maybe Muscheler sees himself as having strengthened the grounds for the rapid-equilibrium assumption, but that assumption is fundamentally flawed. It can't be saved by a marginal adjustment of the forcing in question.

Remember the hypothesis Muscheler is trying to dismiss: that solar activity does have a substantial forcing effect, strong enough to be responsible for late 20th century warming. But if the forcing effect of solar activity is substantial then there is no reason to think that the oceans must have equilibrated to a sustained high level of such forcing by any particular 20th century date, hence no reason to say that late 20th century warming couldn't have been caused by the continuing high level of solar activity.

Perhaps Dr. Muscheler has some other argument for why a steady high level of forcing can't cause warming but if he has been carelessly making the same unstated rapid-equilibrium assumption as Lockwood et al., here is an opportunity to reconsider. We all make unconscious assumptions. Progress in understanding often comes from uncovering and scrutinizing those hidden assumptions, allowing any errors they contain to be corrected. There is no shame in such a re-evaluation. It is how we move forward.

If Dr. Muscheler would like to give a response that is not framed by my commentary I am sure that Anthony would be glad to offer him a guest post. Raimund been game so far, and hopefully will continue to be forthcoming.


My own summary conclusion

There is no possible way to sustain the claim that a steady high level of forcing can't cause continued warming, or to sustain with any confidence the hidden claim that the oceans must have equilibrated to high 20th century solar activity by 1980. Without these claims AR5 goes straight to the trash bin and solar activity is still very much in play as an explanation for late 20th century warming.

If solar activity is responsible for any substantial chunk of that warming then CO2 becomes utterly benign. The IPCC's high estimates of climate sensitivity, needed in order to attribute all recent warming to CO2, are off the table, meaning no possibility of any kind of run-away warming, and if solar activity is the primary explanation for late 20th century warming then the danger going forward is global cooling (now that the sun has turned quiet), making expensive efforts to reduce CO2 emissions the sheerest lunacy.