Guest Post by Willis Eschenbach
In a recent interchange over at Joanne Nova’s always interesting blog, I’d said that the slow changes in the sun have little effect on temperature. Someone asked me, well, what about the cold temperatures during the Maunder and Dalton sunspot minima? And I thought … hey, what about them? I realized that like everyone else, up until now I’ve just accepted the idea of cold temperatures being a result of the solar minima as an article of faith … but I’d never actually looked at the data. And in any case, I thought, what temperature data would we have for the Maunder sunspot minimum, which lasted from 1645 to 1715? So … I went back to the original sources, which as always is a very interesting ride, and I learned a lot.
It turns out that this strong association of sunspot minima and temperature is a fairly recent development. Modern interest in the Maunder sunspot minimum was sparked by John Eddy’s 1976 publication of a paper in Science entitled “The Maunder Minimum”. In that paper, Eddy briefly discusses the question of the relationship between the Maunder sunspot minimum and the global temperature, viz:
The coincidence of Maunder’s “prolonged solar minimum” with the coldest excursion of the “Little Ice Age” has been noted by many who have looked at the possible relations between the sun and terrestrial climate (73). A lasting tree-ring anomaly which spans the same period has been cited as evidence of a concurrent drought in the American Southwest (68, 74). There is also a nearly 1 : 1 agreement in sense and time between major excursions in world temperature (as best they are known) and the earlier excursions of the envelope of solar behavior in the record of 14C, particularly when a 14C lag time is allowed for: the Sporer Minimum of the 16th century is coincident with the other severe temperature dip of the Little Ice Age, and the Grand Maximum coincides with the “medieval Climatic Optimum” of the 11th through 13th centuries (75, 76). These coincidences suggest a possible relationship between the overall envelope of the curve of solar activity and terrestrial climate in which the 11-year solar cycle may be effectively filtered out or simply unrelated to the problem. The mechanism of this solar effect on climate may be the simple one of ponderous long-term changes of small amount in the total radiative output of the sun, or solar constant. These long-term drifts in solar radiation may modulate the envelope of the solar cycle through the solar dynamo to produce the observed long-term trends in solar activity. The continuity, or phase, of the 11-year cycle would be independent of this slow, radiative change, but the amplitude could be controlled by it. According to this interpretation, the cyclic coming and going of sunspots would have little effect on the output of solar radiation, or presumably on weather, but the long-term envelope of sunspot activity carries the indelible signature of slow changes in solar radiation which surely affect our climate (77). [see paper for references]
Now, I have to confess, that all struck me as very weak, with more “suggest” and “maybe” and “could” than I prefer in my science. So I thought I’d look to see where he was getting the temperature data to support his claims. It turns out that he was basing his opinion of the temperature during the Maunder minimum on a climate index from H. H. Lamb, viz:
The Little Ice Age lasted roughly from 1430 to 1850 … if we take H. H. Lamb’s index of Paris London Winter Severity as a global indicator.
After some searching, I found the noted climatologist H. H. Lamb’s England winter severity index in his 1965 paper The Early Medieval Warm Epoch And Its Sequel. He doesn’t give the values for his index, but I digitized his graph. Here are Lamb’s results, showing the winter severity in England. Lower values mean more severe winters.
So let me pose you a small puzzle. Knowing that Eddy is basing his claims about a cold Maunder minimum on Lamb’s winter severity index … where in Lamb’s winter severity index would you say that we would find the Maunder and Dalton minima? …
As you can see, there is a reasonable variety in the severity of the winters in England. However, it is not immediately apparent just where in there we might find the Maunder and Dalton minima, although there are several clear possibilities. So to move the discussion along, let me reveal where they are:
As we might expect, the Maunder minimum is the coldest part of the record. The Dalton minimum is also cold, but not as cold as the Maunder minimum, again as we’d expect. Both of them have warmer periods both before and after the minima, illustrating the effect of the sun on the … on the … hang on … hmmm, that doesn’t look right … let me check my figures …
Well, imagine that. I forgot to divide by the square root of minus one, so I got the dates kinda mixed up, and I put both the Maunder and the Dalton 220 years early … here are the actual dates of the solar minima shown in Lamb’s winter severity index.
Figure 3. H.H. Lamb’s England winter severity index, 1100-1950, overlaid with the actual dates of the four solar minima ascribed to that period. Values are decadal averages 1100-1110,1110-1120, etc., and are centered on the decade.
As you can see …
• The cooling during the Wolf minimum is indistinguishable from the two immediately previous episodes of cooling, none of which get much below the overall average.
• The temperature during the Sporer minimum is warmer than the temperature before and after the minimum.
• The coldest and second coldest decades in the record were not associated with solar minima.
• The fastest cooling in the record, from the 1425 decade to the 1435 decade, also was not associated with a solar minimum.
• Contrary to what we’d expect, the Maunder minimum warmed from start to finish.
• The Dalton minimum is unremarkable in any manner other than being warmer than the decade before the start and the decade after the end of the minimum. Oh, and like the Maunder, it also warmed steadily over the period of the minimum.
Urk … that’s what Eddy based his claims on. Not impressed.
Let me digress with a bit of history. I began this solar expedition over a decade ago thinking, along with many others, that as they say, “It’s the sun, stupid!”. I, and many other people, took it as an unquestioned and unexamined “fact” that the small variations of the sun, both the 11-year cycles and the solar minima, had a discernible effect on the temperature. As a result, I spent endless hours investigating things like the barycentric movement of the sun. I went so far as to write a spreadsheet to calculate the barycentric movement for any period of history, and compared those results to the temperatures.
But the more I looked, the less I found. So I started looking at the various papers claiming that the 11-year cycle was visible in various climate datasets … still nothing. To date, I’ve written up and posted the results of my search for the 11-year cycle in global sea levels, the Central England Temperature record, sea surface temperatures, tropospheric temperatures, global surface temperatures, rainfall amounts, the Armagh Observatory temperatures, the Armagh Observatory daily temperature ranges, river flows, individual tidal stations, solar wind, the 10Beryllium ice core data, and some others I’ve forgotten … nothing.
Not one of them shows any significant 11-year cycle.
And now, for the first time I’m looking at temperature effects of the solar minima … and I’m in the same boat. The more I look, the less I find.
However, we do have some actual observational evidence for the time period of the most recent of the minima, the Dalton minimum, because the Berkeley Earth temperature record goes back to 1750. And while the record is fragmentary and based on a small number of stations, it’s the best we have, and it is likely quite good for comparison of nearby decades. In any case, here are those results:
Once again, the data absolutely doesn’t support the idea of the sun ruling the temperature. IF the sun indeed caused the variations during the Dalton minimum, it first made the temperature rise, then fall, then rise again to where it started … sorry, but that doesn’t look anything like what we’d expect. For example, if the low spot around 1815 is caused by low solar input, then why does the temperature start rising then, and rise steadily until the end of the Dalton minimum, while the solar input is not rising at all?
So once again, I can’t find evidence to support the theory. As a result, I will throw the question open to the adherents of the theory … what, in your estimation, is the one best piece of temperature evidence that shows that the solar minima cause cold spells?
Now, a few caveats. First, I want to enlist your knowledge and wisdom in the search, so please just give me your one best shot. I’m not interested in someone dumping the results of a google search for “Maunder” on my desk. I want to know what YOU think is the very best evidence that solar minima cause global cooling.
Next, don’t bother saying “the Little Ice Age is the best evidence”. Yes, the Maunder occurred during the Little Ice Age (LIA). But the Lamb index says that the temperature warmed from the start of the Maunder until the end. Neither the Maunder’s location, which was quite late in the LIA, nor the warming Lamb shows from the start to the end of the Maunder, support the idea that the sun caused the LIA cooling.
Next, please don’t fall into the trap of considering climate model results as data. The problem, as I have shown in a number of posts, is that the global temperature outputs of the modern crop of climate models are nothing but linear transforms of their inputs. And since the models include solar variations among their inputs, those solar variations will indeed appear in the model outputs. If you think that is evidence for solar forcing of temperature … well, this is not the thread for you. So no climate model results, please.
So … what do you think is the one very best piece of evidence that the solar minima actually do affect the temperature, the evidence that you’d stand behind and defend?
My regards to you all,
[UPDATE] In the comments, someone said that the Central England Temperature record shows the cooling effects of the solar minima … I’m not finding it:
As you can see, there is very little support for the “solar minima cause cool temperatures” hypothesis in the CET. Just as in the Lamb winter severity data and the Berkeley Earth data, during both the Dalton and Maunder minima we see the temperature WARMING for the last part of the solar minimum. IF the cause is in fact a solar slump … then why would the earth warm up while the sun is still slumping? And in particular, in the CET the Dalton minimum ends up quite a bit warmer than it started … how on earth does this support the “solar slump” claim, that at the end of the Dalton minimum it’s warmer than at the start?
The Usual Request: I know this almost never happens, but if you disagree with something that I or someone else has said, please have the common courtesy to QUOTE THEIR EXACT WORDS that you disagree with. This prevents much confusion and misunderstanding.
Data: Eddy’s paper, The Maunder Minimum
Lamb’s paper, The Early Medieval Warm Epoch And Its Sequel
Berkeley Earth, land temperature anomalies