Guest Post by Willis Eschenbach

My mind runs to curious corners of the climate question. I got to thinking about how little the atmospheric CO2 level has changed over most of the last two thousand years. Here is the CO2 data from ice cores (colored dots) and from the Mauna Loa CO2 measuring stations (red/black line).

Figure 1. Ice core CO2, and Mauna Loa instrumentally-measured CO2.

Note that there is good agreement between the thirteen different ice cores, as well as good agreement during the period of the overlap between the ice core and the instrumental CO2 data.

Looking to see how the CO2 related to the temperature since the year dot, I took that CO2 record and I overlaid it on the Ljundqvist reconstruction of the temperature variability of the extratropical northern hemisphere from 30°N to 90°N. It is a decadal average record that goes from the year 1 to the year 1999. I also added in the recent Berkeley Earth decadal averages for the period since the end of the study, for 2000-2009, and 2010-2019. (Because one is a proxy record and the other is an instrumental record, I have variance adjusted the Berkeley Earth record to match the variance of the proxy record over the period of their overlap, which is from 1850 to 1999. This is the same procedure that Ljundqvist used for the black dashed line in his reconstruction.)

Here is that graphic.

Figure 2. Ljungqvist decadal temperature reconstruction 90-30°N, to which I’ve added ice core and instrumental CO2, as well as the two final decades of Berkeley Earth mean 90-30°N temperatures (blue circles). The ice core data is a lowess smooth of the individual ice core data points shown in Figure 1.

In Figure 2, we can see the early “Roman Warm Period”, which lasted up to about the year 200 AD. Then temperatures dropped until about the year 500 AD. After that, the world warmed up again to the “Medieval Warm Period”. Then the temperatures cooled to the “Little Ice Age” which ended about the year 1700 AD, and finally, temperatures have warmed in fits and starts for the three centuries since the Little Ice Age.

Here are my questions about this historical temperature record:

• Why did the temperature start dropping after the Roman Warm Period? Why didn’t it just stay warm?

• Why did the cooling start in 200 AD, and not say in the year 600 AD?

• Why did the temperatures start warming around 550 AD, and continue warming up to the Medieval Warm Period peak at around 1000 AD? It could have stayed cold … but it didn’t.

• Why was that warming from 550 to 1000 AD, and not from say 800 to 1300 AD?

• What caused the steady cooling from about 1000 AD to the depths of the Little Ice Age, where temperatures bottomed out around 1700 AD?

• Why was that cooling from about 1000-1700 AD, and not e.g. 1250-1850 AD?

• Instead of stopping at the year 1700 AD, why didn’t the world keep cooling down to real glaciation? Given the Milankovich cycles and the lengths of the other warm interglacial periods, we’re overdue for another real ice age.

• Why did temperatures start warming start again at the end of the Little Ice Age, instead of just staying at the 1700 AD temperature?

• Why has it warmed, in fits and starts, from the Little Ice Age up to the present?

Here’s why all of that is important.


Not one.

And from Figure 2 above, it is quite clear that the answer is not “CO2” …

Without having enough understanding of the climate to know the answers to those questions, there is absolutely no way to say that the recent warming is not merely more natural fluctuations in the earth’s temperature.

Speaking of fluctuations, I note in Figure 2 that the Earth’s extratropical northern hemisphere surface temperature has wandered around in a range of about 1.5°C over the last 2000 years. The earth’s surface temperature is on the order of 288 kelvin, so including the variance adjustment, this represents a temperature variation on the order of one percent over a two-thousand-year period … to me, this is the most interesting thing about the climate—not how much it changes over time, but how little it changes.

This thermal stability is not from thermal inertia—the land in each hemisphere swings about 15°C over the course of every year, and the ocean in each hemisphere swings about 5°C every year.

I say that this surprising and exceptional stability is due to the thermoregulatory action of emergent climate phenomena including dust devils, tropical cumulus fields and thunderstorms, the El Nino/La Nina mechanism that pumps excess heat to the poles, tornadoes, and hurricanes. I describe this theory of emergent thermoregulation in a series of posts listed in my 2021 index of my posts here in the section called “Emergence”. These act to keep the earth from either excessive warming or cooling.

Anyhow, that’s where my mind wandered to most recently, to the question of two thousand years of temperature and how it’s not ruled by CO2 …

Today was sunny and the ground was dry for the first time in a while, so I spent far too much of the afternoon underneath my truck unbolting and taking off the camper hold-down brackets. There’s nothing like discussing the concept of excess rotational inertia with a recalcitrant bolt, a bolt located where you can only get one hand in to twist the wrench, to give a man a reminder of his lack of importance in this miraculous world of wonders … made me glad I own an impact wrench, but there are places where it can’t go.

But as the I Ching says, “Perseverance furthers”, and after using end wrenches, sockets, breaker bars, deep-wall sockets, two impact wrenches, and saying more than a few bad words, I got it all done.

Best of the sunshine to everyone, life is good,


Technical Note: Some folks have asked, is it OK to splice proxy and instrumental data, or more generally, any two datasets purportedly measuring the same phenomenon?

My answer is absolutely yes, IF two conditions are met:

Condition 1: There is enough temporal overlap between the datasets to verify that the two datasets are giving the same answer. In this case, for example, there is an overlap between instrumental and proxy data for both the ice core/instrumental CO2 and the proxy/instrumental temperature records. Figure 1 shows the ice-core/Mauna Loa CO2 overlap. Here is a closeup of the overlap between the Ljungqvist temperature and the Berkeley Earth variance-adjusted temperature.

Figure 3. Closeup, overlap of Ljungqvist temperature and variance-adjusted Berkeley Earth temperature

Looking at Figures 1 and 3, you can see the overlap in each case and verify that the proxy and instrumental records are clearly in good agreement.

Condition 2: The fact of the splice is made clear and evident, with no attempt to hide it. In both graphs, the overlap period is clearly shown, and the existence of the splice is acknowledged in the text as well

Here’s a post of mine which discusses splicing, appropriately entitled “Splicing Clouds” …

Usual Note: I can defend my own words. I can’t defend your interpretation of my words. So when you comment, please quote the exact words that you are referring to.

via Watts Up With That?

March 22, 2021 at 12:27PM