On climate change and the Sun

About a month and a half ago I did a post called “What is Truth?” This is not a direct continuation but more of a sequel. There I pointed out how statements appear in the press, or the news, or echoed out by some pundit as being true, that a trained scientist will immediately question. And since Truth (Biblical or scientific) cannot be a relative thing, cannot be a “matter of opinion”, I showed how we can all test the veracity of such claims, and settle the matter. All it takes is consulting the peer reviewed literature.

An article appeared today (October 26,2022) in the Arizona Republic, entitled “How do we know humans triggered climate change?” In it, the following statements are made:

One way to show humans caused the warming “is by eliminating everything else”, said Princeton University climate scientist Gabe Vecchi…They measure what they call radiative forcing in watts per meter squared.

The first most frequent natural suspect is the sun. The sun is what warms the Earth in general providing about 1,361 watts per meter squared heat, year in year out. That’s the baseline, the delicate balance that makes Earth livable. Changes in energy coming from the sun have been minimal, about one-tenth of a watt per meter squared, scientists calculate...

You got that, right? He claims there is only a 0.1 W/m2 variation in what is called the Total Solar Irradiance (TSI).

But what does the literature say?

Let’s look at an actual plot of TSI from the year 1750 through 1990. This historical data can be reconstructed because good correlations have been found between sunspot number, solar irradiance, and 10Be and 14C production in the atmosphere. Sunspot numbers have been recorded since Galileo invented the telescope, while the historical atmospheric isotope concentrations can be deduced from samples of deep ice cores. Taking all those (and other) factors into consideration, Hoyt and Schatten show this result (Figure 1) where the TSI (dark solid line) is compared to the Northern Hemisphere temperature variation over the same period (lighter dashed line):

Figure 1. From Hoyt and Schatten 1997, comparing TSI to Northern Hemisphere temperature variation over two and a half centuries.

(Figure 1 here is Figure 10.21 from The Role of the Sun in Climate Change; by Douglas V. Hoyt , and Kenneth H. Schatten; page 195, Oxford University Press, Incorporated; 1997-04-03. That book is available on line if you would like to read it yourself. The link takes you to chapter 3).

We see a TSI varying from a low of 1368 W/m2 to a high of about 1372 W/m2. That is a variation of +/- 2 W/m2 or 20 times larger than the article cites. Yep, that is an error in the newspaper article of more than an order of magnitude.

Was that a typo, or is someone lying?

Since the publication of Hoyt and Schatten’s book, researchers have been developing sophisticated computer models of the sun itself and comparing their predicted TSI to recent directly measured data to calibrate the models.

Thus, in W. K. Schmutz’s “Changes in Total Solar Irradiance and climatic effects”, J. Space Weather Space Clim. 2021, 11, 40; published by EDP Sciences 2021; the author uses the results of one of those models to make the following comparison.

Figure 2. From W. K. Schmutz, 2021, showing computed Sun model TSI over 6 centuries compared to recorded temperature data in Kyoto bay, Japan over the same time period.

The red curve in Figure 2 is the TSI data from Egorova et al, “Revised historical solar irradiance forcing”; Astron Astrophys 615:A85. The dashed lines are temperature data from Kyoto bay over the same time.

The TSI absolute scale in Figure 2 is different from that of Figure 1 simply because the two models use different assumptions, and most likely due to different “calibration” data. The Egorova et al result is a modern computer model of the spectral solar irradiance. As they explain in their abstract: Their TSI is “calculated as a weighted sum of the contributions from sunspot umbra, sunspot penumbra, faculae, and quiet Sun, which are pre-calculated with the NLTE Spectral SYnthesis code (NESSY)… and a new structure model for the quietest state of the Sun.” Whereas Hoyt and Schatten are reporting on an empirical model using other measured data beside the solar activity.

Regardless of the difference in the absolute TSI scale between Figure 2 and Figure 1, these authors also show oscillations of the order of +/- 2 W/m2 about a varying mean. More importantly, although we cannot expect mean Northern Hemisphere temperature and the Kyoto bay Temperature in Japan to be exactly the same, the fact is inescapable that TSI variations of +/- 2 W/m2 relative to a running mean do correlate to the long time oscillations of the temperature of the Earth.

Note that there is not only one way to plot a comparison as that shown in Figure 2 because the two phenomena (Earth temperature and TSI) are two different functions measured on two different scales with different units, each one dependent on many variables. We can say there is a correlation; but to make that correlation quantitative we would have to develop a model that connects the two and includes the other variables that affect the temperature of the Earth beside TSI. Nevertheless the correlation is there.

To see it better, I have roughly traced over the Temperature data represented by the green dotted curve and the TSI data and recreated the Figure with the Temperature axis compressed and shifted, because, as it is, we do not have a first principles absolute correlation between these axes. In other words it was the authors’ choice how they made that plot. TSI is in red, Temperature is dashed green:

Figure 2b, Rudy’s redrawing of Figure 2 to highlight the correlation between TSI and Earth Temperature.

The Figure above emphasizes that the variation in TSI over long time scales does appear to correlate with long time variations of the Earth’s temperature. Furthermore, it points out that if we focus on the upwards swing of the temperature data away from the TSI trend, that we see past 1930, and we choose to blame it all on human and Industrial revolution effects, we would have to then explain what caused a similar deviation before the year 1500.

But, I digress.

We see above that peer reviewed literature data from 1997, 2018, and 2021 all agree that the Total Solar Irradiance has long term variations of the order of +/- 2 W/m2 not the 0.1 W/m2 that Dr. Vecchi gives in the Arizona Republic article.

Then where does Dr. Vecchi get his number?

In an article entitled “Explainer: Why the sun is not responsible for recent climate change” by Zeke Hausfather of carbonbrief.org, the following graph is shown, where the red curve only shows +/- 0.5 W/m2 oscillations. This is much lower than the long time oscillations visible in the previous two graphs! What gives?

Figure 3. Hausfather’s web article comparison of TSI to variation in global mean temperature over 1.4 centuries.

Hausfather’s data resembles data found on Nasa’ s global climate change website based on SATIRE-T2.

Why the big difference? Is someone lying?

Well, SATIRE-T2 is a surface flux transport model (SFTM) that describes the evolution of the solar surface magnetic field using as its sole inputs the sunspot number and total sunspot area. This is not a full sun model as Rogarova et al developed, nor is it an empirical model as that reported by Hoyt and Schatten using, among other things, 10Be and 14C atmospheric isotope data. In other words the SATIRE-2 data is incomplete. This is well known: Rogarova et al, highlight the difference between their model and SATIRE model in this plot:

Figure 4. From Rogarova et al comparing the TSI calculation form their full sun model to the sunspot-only SATIRE model.

HUGE difference: The SATIRE model is the green (and blue) mostly flat data at the top of the graph that only highlights the small ripple oscillations associated with the well-known 11 year mean sun cycle. By contrast, in orange, aqua, red, and violet, we see 4 results based on slightly different input parameters of the full sun model.

Clearly TSI, according to the more complete model, has swung from 1355 to as high 1361.5 W/m2 in the past and more recently has shown swings from 1357.5 to 1361. The bottom line is that +/- 2 W/m2 variations from the local mean is what the full Sun model predicts.

The important point to see is that Vecchi and Hausfather, whose articles have the avowed purpose to show us that the sun is not the cause of large scale climatic temperature change only report on the SATIRE data. But even if they did that correctly they would have to tell you that those small ripples are +/- 0.5 W/m2, a number 5 times larger than the 0.1 figure that Vecchi cited in the newspaper article.

To get anywhere close to 0.1 you have to use the smoothed running average line (the dashed red curve of Figure 3).

And if you think that is a useful, honest measure, let me point out that you do not need ever to cover your plants and vegetables in the winter here in Phoenix because even if the temperature occasionally drops to 30F at night, it is going to be about 60F at worst during the day, and the mean of 60F and 30F is 45F – certainly not freezing.

Who is lying?

The newspaper article’s number is what results when you average out the ripples of the incomplete model that misses the bulk of the long-time variations of the incoming power from the sun. Was this an oversight? an innocent mistake? We are justified in being suspicious when we notice that both Vecchi’s and Hausfather’s avowed goal is to convince us that variation in the energy output from the Sun is not a significant contributor to climate change. And, lo and behold, their “scientific” data proves them right!

You can reach your own conclusion…

My point in this blog post was not to tell you that climate change is not happening. Almost by definition, climate has to change. It depends on so many variables that it would be utterly astounding if it were consistent and predictable. Certainly, human beings and the changes we make on the environment must have effects in the short and long term. Many of those effects are to be expected from basic Chemistry and Physics. But it does not help anybody when supposed “scientific” data is presented as fact when in reality it is rhetoric.

I guess this is not a happy conclusion for the layman; but it is a truth we must accept:

If you want to understand what Science says, you need to follow the rules of the scientific method and of correct scientific practice. Do not just accept the propaganda that you happen to like.

As a minimum what you have witnessed in the AZ Republic article is sloppy scientific reporting. But, like I have said before, scientists know that scientists can be biased, like all human beings. The good news is that the quickest, surest way to prove some statement wrong is to look at the literature. That record is there and it is not going to go away. Truth is not relative.

Share this on:


Sign up to receive new stories in your email as they’re published.

Your privacy is important. We won’t send spam or share your email address. Privacy Policy

Leave a Reply

Your email address will not be published. Required fields are marked *