The Paper that Blew it Up

By Andy May

“If you can’t dazzle them with brilliance, baffle them with Bull…” W. C. Fields

In late February 2015, Willie Soon was accused in a front-page New York Times article by Kert Davies (Gillis & Schwartz, 2015) of failing to disclose conflicts of interest in his academic journal articles. It isn’t mentioned in the Gillis and Schwartz article, but the timing suggests that a Science Bulletin article, “Why Models run hot: results from an irreducibly simple climate model” (Monckton, Soon, Legates, & Briggs, 2015) was Davies’ concern. We will abbreviate this paper as MSLB15. Besides Soon, the other authors of the paper are Christopher Monckton (senior author, Lord Monckton, Viscount of Brenchley), David Legates (Professor of Geography and Climatology, University of Delaware), and William Briggs (Mathematician and statistician, former professor of statistics at Cornell Medical School). In the January 2015 article, the authors “declare that they have no conflict of interest.”

MSLB15 was instantly popular and devastating to the climate alarmist cause and to the IPCC Fifth Assessment Report (IPCC, 2013). The IPCC is the Intergovernmental Panel on Climate Change, a research organization set up by the United Nations in 1988. MSLB15 was published online January 8, 2015 and downloaded 22,000 times in less than two months, an outstanding number of downloads. The New York Times article appeared less than two months after MSLB15 hit the internet, it was a “fake news hit job.”

The paper caused a stir because it explained that the IPCC’s Fifth Assessment Report or “AR5” reduced its near-term warming projections substantially, but left its long-term, higher, projections alone. This was because the IPCC central, CO2 feedback-based, estimate of the climate sensitivity to CO2 was reduced from 3.2°C (5.8°F) to 2.2°C (4°F) per doubling of CO2 concentration. The sensitivity to CO2 is often abbreviated “ECS” for Equilibrium Climate Sensitivity. The MSLB15 calculation was done the way the IPCC used in their Fourth Assessment Report, abbreviated “AR4.”

If the new estimate is correct, the projected rise in temperature for the 21st century is less than one-degree C. Another implication of the change is that the combustion of all fossil fuels estimated to exist would only cause a temperature increase of 2.2°C (4°F). This amount of warming is trivial, good for humanity, but bad for the climate alarmists.

The organization that models climate projections for the IPCC is the CMIP, or the Coupled Model Intercomparison Project. It was created in 1995 to consolidate climate models from around the world into a set of projections that could form the basis for the IPCC reports. The CMIP climate models used for the IPCC fourth and fifth assessment reports overestimate global warming by a substantial amount as shown in John Christy’s plot from a previous post and shown again here as Figure 1.

Figure 1. John Christy’s famous graph comparing the AR5 IPCC climate models to weather balloon and satellite observations for the mid-troposphere. The satellite and weather balloon observations are independent of one another and surface measurements. From Christy’s 2016 Congressional testimony (Christy, 2016).

AR5 was essentially a repeat of AR4 with respect to the computation of human influence on climate. Yet, MSLB15 tells us that deep in AR5 a dramatic change was made in the model calculations that lowers the computed climate impact of CO2. But the change was not reflected in the AR5 long-term climate projections. Monckton points out that the IPCC made the changes due to pressure from expert reviewers to bring their climate projections and model parameters into line with observations (Monckton, 2015b). The IPCC made the change, then ignored it in their longer-term projections.

Modern computer climate models are expensive “general circulation” models that model thermal energy moving through the atmosphere and the upper part of the oceans. The models break the atmosphere into 3D grid boxes that are assumed to be in local thermodynamic equilibrium and only change at their edges where they contact neighboring boxes. The older models, such as the 1979 Charney model (Charney, et al., 1979), were simpler and modeled the whole atmosphere and upper ocean conceptually.

As discussed in our last post, the complexity of modern models has not changed the estimated climate sensitivity to CO2 or made it more accurate. The 1979 Charney Report model computed the same range of sensitivity to CO2 as AR5 reported in 2013. This range (1.5° to 4.5°C) has survived intact for forty years despite the efforts of thousands of researchers spending over one-hundred billion U.S. 2014 dollars between 1993 and 2015 (U.S. Government Accountability Office (GAO), 2016) in the U.S. alone.

So, when MSLB15 showed up online, explaining that the AR5 model’s feedback estimates suggested an ECS of 2.2°C (4°F), rather than the AR4 estimate of 3.26°C (5.9°F) (IPCC, 2007, p. 798) it caused a huge uproar. As Rud Istvan noted in a post, at the time, “If you are taking heavy flak, you are over the target.” The B-17 or Avro Lancaster being flown by Christopher Monckton, Willie Soon, David Legates and William Briggs must have been directly over central Berlin given the response by the alarmists and the news media.

The direct warming from CO2 or ECS is small, around one-degree Celsius for a doubling of CO2. This slight warming will cause a feedback, generally assumed to be due to an increase in absolute humidity, caused by warmer temperatures. Water vapor is a much stronger greenhouse gas than CO2, so this reduction in feedback, from AR4 to AR5, is what the climate alarmists are worried about. Why is the range of ECS in AR5 the same as in AR4 when such an important component of CO2-caused warming was reduced? Did politics overrule the scientific findings?

Adding fuel to the fire was the fact that no best estimate of ECS was given in AR5. There are many ways to compute ECS and they disagree so much, that the IPCC did not give a best estimate. Both TAR (the IPCC third assessment report) and AR4 provided a best estimate of 3°C (5.4°F), so if AR5 had stated their feedback-implied ECS of 2.2°C (4°F), the precipitous decline would have been obvious and politically damaging. So, they were silent. The obvious question is why? Did they think no one would notice the intellectual dishonesty?

To estimate ECS, one can use climate model results, analysis of feedbacks (like in AR4 or MSLB15), observed temperature and CO2 changes (Lewis & Curry, 2018), or paleoclimate studies. The dilemma the IPCC faced in AR5 was that these estimates did not agree and many of them were far below those given in AR4 and previous assessment reports, as shown in our previous post. One wonders why the IPCC is so sure that humans control the climate with their greenhouse gas emissions, when the impact of the main greenhouse gas, CO2, is so poorly understood? Since no best estimate of ECS was given in AR5, one can argue that our understanding is diminishing with time.

Once Christopher Monckton and his co-authors, including Willie Soon, noticed that the CO2 feedback forcing was lowered in AR5, they created a simple model to investigate this difference and published their assessment. It is virtually impossible to attack the “Irreducibly simple climate model” presented in the paper, it is too basic. As Istvan reports the derivation of the MSLB15 model is impeccable. So, the alarmist cabal initially said that Science Bulletin was an obscure journal, therefore the paper cannot be any good. Predictably, that didn’t work, besides, the Science Bulletin is the Chinese version of Nature or Science.

Criticism of MSLB15
Rud Istvan’s post on the paper is illuminating and interesting, as is Monckton’s reply. Many traditional climate scientists, even Judith Curry, are somewhat dismissive of MSLB15. They think this simple approach to climate modeling doesn’t provide any insights into why the climate models do not agree with observations. Kevin Trenberth complains that the model is too simple (Briggs, 2015). Istvan comments that: “Trenberth’s comments to the NYTimes are indefensibly misleading in my opinion, and provide a vivid object lesson about consensus climate ‘science’ and its reporting” (Istvan, 2015). We agree with this assessment. MSLB15 explicitly recognize that their model is simple:

“[The MSLB15 model] is not, of course, intended to replace the far more complex general-circulation models; rather, it is intended to illuminate them.” (Monckton, Soon, Legates, & Briggs, 2015)

The irreducibly simple model is simple, it is in the title of the paper and Trenberth’s statement to the New York Times (Briggs, 2015) is vacuous. MSLB15 is important, not as an advance in climate science, but because it illuminates the serious flaws and internal contradictions in the IPCC/CMIP climate models. Further evidence that the IPCC models are seriously flawed is that they are no more accurate in predicting the climatic impact of CO2 now than they were in 1979, the MSLB15 model merely drives this painfully obvious point home. Billions have been spent; one would think we would have seen some progress by now.

The subtitle of this post, a quote from W. C. Fields, “If you can’t dazzle them with brilliance, baffle them with Bull…” says it perfectly. The IPCC computer models and the ludicrous idea that averaging them provides us with a reliable and useful prediction of future climate is an attempt to “baffle us with bull….” This human-caused climate change perpetual money-squandering machine must start producing answers or be cut off from funding.

The MSLB15 model reduces the nonsense to its essence and shows this deception, if not clearly as we would like, at least more clearly and succinctly than the IPCC does. Compared to the real world, the IPCC models are too simple, their complexity doesn’t help us understand the human impact on climate, it merely provides a way to hide their inadequacies and push a chosen agenda. This was what I took away from reading MSLB15.

Rud Istvan thinks the simple model could be made simpler and have the same effect. Monckton thinks the model needs the all the elements it has, to be useful. Either way, Istvan found the model to be useful and we agree. I have no problem with the model as a useful way to understand the more complicated general circulation models. It is not, as MSLB15 readily admits, a replacement for them. It sheds light on them and provides a useful reality check.

The point MSLB15 makes, is that the IPCC model based ECS estimates are inflated. They could add that they are inaccurate and are not improving with time and money spent. Monckton says in his rebuttal to Istvan, we must let “the daylight in on the magic” (Monckton, 2015b, p. 6). We agree.

Mark Richardson and colleagues (Richardson, Hausfather, Nuccitelli, Rice, & Abraham, 2015) try to show that the MSLB15 model underestimates global mean temperatures. Richardson, et al. do not refute MSLB15, they simply refute a strawman of their own creation. Further, the only period that Richardson, et al. use, that is long enough to be considered “climate,” is 1900 to 2010. For this period, both CMIP5 and MSLB15 have errors that are well within the margin of error for the temperature datasets they cite, HadCRUT4, Cowtan & Way, and Berkeley Earth. Their shorter periods, 1970-2010 and 2000-2010 are too short to be meaningful.

Next, the alarmists, possibly including John Holdren, senior advisor to President Obama, began to attack Willie Soon, one of the authors, through his employer, the Harvard-Smithsonian Center for Astrophysics. John Holdren had already attacked Willie Soon and Sallie Baliunas’ 2003 papers (Soon & Baliunas, 2003) and (Soon, Baliunas, Idso, Idso, & Legates, 2003b) when he was still at Harvard according to The Harvard Crimson (Sanchez, 2003). He claimed the papers were a “flawed analysis.” They were not flawed and MSLB15 was not flawed either. MSLB15 might be overly long and a difficult read, but it is not flawed, as far as we can tell.

Unable to attack the science, the alarmists wanted the skeptics in the Harvard-Smithsonian Center for Astrophysics silenced. The Smithsonian responded with new directives on conduct that contained a “loyalty to the Smithsonian” clause. The Smithsonian’s Inspector General investigated Soon and found no wrongdoing on his part, but this simply enraged the critics and didn’t settle anything (Arnold, 2016). Attacks on climate skeptics were common in 2015 and 2016 and the Obama administration was not alone, some of the harassment came from Congress, particularly from Senator Sheldon Whitehouse and Congressman Raúl Grijalva.

The New York Times and the other news organizations covering the story should have written about what MSLB15 said, the story isn’t that complicated or hard to explain. But they didn’t. The fact they attacked the authors, without discussing what they wrote in their peer-reviewed paper, speaks volumes, as stated in the web site “Bishop Hill” by Andrew Montford (Montford, 2015). The news media didn’t care about climate science, after all, the “science is settled,” isn’t it?

The 2015 Attack
As mentioned at the top of the post, the height of the attacks on Willie Soon, by the New York Times (Gillis & Schwartz, 2015) was on February 21, 2015. They attacked Willie Soon personally. They relied upon false information from Kert Davies (Davies, 2020), the founder of the secretive Climate Investigations Center or CIC. Davies suggested that Willie Soon had a conflict of interest and lied in MSLB15 when he said he didn’t. Davies and the New York Times claimed that Soon had received undisclosed money from ExxonMobil and the Southern Company.

Most of the New York Times article is either wrong or misleading and in our new book, Politics and Climate Change: A History, we address each of their accusations. Here we will just cover a few of the most egregious lies. The basis for the attack was a Freedom of Information Act request (FOIA) to obtain internal documents from the Harvard-Smithsonian Center for Astrophysics, where Soon is employed as an astrophysicist. The FOIA was filed by Davies and Greenpeace.

As he had previously done in 2010 (see our book for details of the 2010 FOIA request), the director, Charles Alcock, made a crucial mistake and ordered Willie Soon to comply with the request. Unlike departments in the Executive branch of the government, a government trust, like the Smithsonian Institution, does not have to comply with FOIA requests. Thus, Alcock’s order is persecution of an employee. Alcock is specifically allowing Davies, the New York Times, and Greenpeace to intimidate and harass one of his employees. The documents (New York Times, 2015) include research proposals from the Smithsonian Astrophysical Observatory that were written by Soon to The Southern Company (NYSE: SO) a leading natural gas and electric utility company, ExxonMobil (NYSE: XOM), the Charles Koch Charitable Foundation, and Donor’s Trust.

Science is a process for challenging the consensus view. Science cannot prove anything, the scientific process is about disproving things, particularly consensus opinions. For example, both Copernicus and Galileo disproved the idea that the Sun revolves around Earth. Science uses observations, analysis, and logic to disprove erroneous assumptions made by the public.

The New York Times obviously does not understand this 9th Grade definition of the scientific method and their article asserts:

“The documents shed light on the role of scientists like Dr. Soon in fostering public debate over whether human activity is causing global warming. The vast majority of experts have concluded that it is, and that greenhouse emissions pose long-term risks to civilization.” (Gillis & Schwartz, 2015)

This unsupported assertion is laughably anti-scientific. As we have seen, “the vast majority” or a consensus of scientists is a political thing. A scientist looks at the conclusion of the “vast majority” and asks, “Is that true? How can I test that idea?” Challenging the consensus view is the whole idea of science. A true scientist wants to foster “public debate.”

The premise of the New York Times article is quite disturbing for several reasons. Firstly, they assume the so-called “consensus” view that climate is controlled by humans is true, even though no direct evidence supporting it exists. The computer model projections relied upon by the IPCC are not direct evidence. In fact, MSLB15 suggests the models are not even accurate. Let us not quibble over the words “causing climate change” and “controlling climate.” Everyone agrees that humans have some influence on climate, the debate is over how much. The alarmists clearly believe that CO2 is the “control knob” for climate change (Lacis, Schmidt, Rind, & Ruedy, 2010).

Secondly, they assume that privately funded research, by an established and very credible astrophysicist, working for the Harvard-Smithsonian Center for Astrophysics, is somehow tainted by donations to the Smithsonian. Thirdly, they seem to think that since Soon “has received little federal research money over the past decade” that this somehow makes him inferior to other researchers. All three assumptions are horrible. Do they really think that private companies should not be allowed to fund scientific research? Or, if they do, that the research should be discounted based only on the source of funding?

These views are not only juvenile, they are anti-scientific and possibly violate the free speech portion of the first amendment of the U.S Constitution. It is illegal to attempt to take away a person’s constitutional rights through intimidation or other means (Columbia Law School, 2020).

One of the Smithsonian studies, partially funded, by ExxonMobil, Donor’s Trust and the Southern Company was “Understanding Solar Variability and Climate Change: Signals from Temperature Records of the United States.” For one interested in climate change this would seem to be an important topic to investigate. The checks from these organizations were made out to the Smithsonian Astrophysical Observatory or the Smithsonian Institution (see my book for photocopies of the checks). No money was paid to Willie Soon, who is a government employee and paid a salary. He wrote the proposals for the Smithsonian Institution as one of his duties as a Smithsonian employee (Arnold, 2016).

Science stands on its own, the conclusions either follow from the evidence and analysis presented, or they do not. The study can be replicated, or it cannot. Funding has nothing to do with it. Just because the New York Times reporters cannot understand Soon’s papers, does not mean no one can. Other scientists will read his papers with a properly skeptical eye and let him, or others, know if there is a problem. The papers survive or fail on their own merits.

The first amendment grants people and through them corporations, the right to free speech and the right to petition the government for a redress of grievances. This concept is supported by the Supreme Court in rulings like Citizens United (Smith, 2020). The New York Times article complains that Soon presented his research, funded through the Smithsonian, by the Southern Company, ExxonMobil and the Donor’s Trust, to Congress. Are they saying that Soon and the people who funded some of his research should have their first amendment rights taken away because they disagree with “most” scientists or the New York Times? That is not the way science, or the United States works. In general, the article was anti-science and anti-American.

The scientific community provides a place for scientists to debate ideas. The scientific playground contains thousands of peer-reviewed journals that allow all sides an opportunity to present their data, analysis, and conclusions for inspection. Unfortunately, once politicians and the news media became involved in the human-caused climate change debate it became a disaster. Politicians used personal attacks, suppression of opposing views, ridicule, harassment, and intimidation, rather than reason to push their views on scientists. All of these were used against Willie Soon and his former supervisor Sallie Baliunas. His friends and colleagues, David Legates, Christopher Monckton, and William Briggs, were also attacked unfairly. Politics and a scientifically ignorant news media corrupt science to an unacceptable degree. We are opposed to all government funding of scientific research for this reason. My next post and my new book discuss this viewpoint further.

This is an abbreviated excerpt, with minor modifications, from my new book, Politics and Climate Change: A History.

To download the bibliography, click here.

I’ve been requested to post a pdf of this post, download it here.