Humlum is at it again

In their opinion article in the widely-read Norwegian newspaper Teknisk Ukeblad  (translation available here) which followed an initial letter in the same newspaper, Dr. Bye, Prof. Humlum, and Dr. Stordahl (BHS), attempt to refute what they call the “CO2 hypothesis” (usually called in the literature Anthropogenic Global Warming, or AGW theory – it is a theory). Their argument is articulated as follows: firstly, they dismiss categorically that a methodology exists which would allow for attributing the current observed global warming to human activities with a confidence of at least 90%.  BHS pretend that the IPCC assigns confidence levels to scientific positions based on the outcome of a simple vote.  They suggest that the more votes a statement (e.g., “global warming is mostly due to human activities”) receives, the higher confidence the IPCC assigns to it. As discussed below, this is by no means how the IPCC determines confidence levels.  BHS are criticizing the IPCC for something it does not actually do.

Having seemingly wiped out the entire AGW hypothesis theory by this ingenious but incorrect argument, BHS then take on the challenge to propose an alternative explanation for the increase in the mean global temperatures observed since the beginning of the Industrial Revolution –an increase that they do indeed accept as real and scientifically valid.  BHS suggest the warming is simply due to natural factors. To support their hypothesis, they rely on a simple figure (incidentally also present on the website of one the authors) showing the evolution of the atmospheric CO2 concentration as well as the simultaneous temperature record taken from the GISP2.  Skeptical Science readers may recall this figure from Rob Honeycutt's Crux of a Core 3 (Figure 1).

Figure 1: GISP2 temperature data and EPICA Dome C CO2 data from

Proud of their achievement, BHS end the correspondence by reminding the reader that they are strong defenders of the true and pure scientific methodology.

Unfortunately for BHS, we feel obliged to notify them that nothing in their communications comes close to such an ideal, unbiased, and intellectually honest scientific approach. This post aims at providing with some clarifications that have been regrettably omitted from the BHS correspondence, and that we feel are essential for the interested reader to take an informed position on the issue of AGW.

Their whole critique is directed towards the attribution of climate change: basically, is it mostly natural (the BHS hypothesis) or mostly anthropogenic (the AGW hypothesis theory)? The Working Group I of the IPCC has dedicated Chapter 9 to this question in its latest assessment report (AR4WG1), to which we refer in the following. We also provide throughout this post links to peer-reviewed scientific papers, or to web pages linking to such publications, because we believe that any claim must be substantiated by relevant and solid supporting evidence.

About the likelihood of AGW

Let us start with the definition of the likelihood [1]. BHS base their work on a single quote coming from a single scientist, Prof. Eystein Jansen.  This is not scientifically rigorous, especially when the quote is taken from a TV show that by nature does not allow for complexity to be discussed in details and where approximations are often tempting to make for the sake of brevity.  BHS should know that scientific methods have never been defined on air but in the peer-reviewed literature.

The IPCC works with likelihoods, summarized in Table 9.4 of Chp. 9. Likelihood “refers to a probabilistic assessment of some well defined outcome having occurred or occurring in the future” [2].

The likelihood for the observed surface warming of the Earth in the last 50 years is very likely (>90%) not to be due to natural causes alone, and very likely (>90%) to be dominantly caused by greenhouse gas forcing [3]. In other words, taking into account all the known factors contributing to defining the behavior of the climate system, and in particular the human-caused emissions of greenhouse gases, the probability of obtaining the current surface temperature (including increased ocean temperatures) is more than 90%.

The method followed to derive these probabilities that define likelihood “may be based on quantitative analysis or an elicitation of expert views” [2]. In addition, whatever the outcome of the estimation process, sources of uncertainty must be mentioned and discussed. The treatment of uncertainties by the IPCC is discussed in Section 1.6 of AR4WG1, and has been the subject of a recent meeting of the panel experts.

So no, the 90% likelihood that global warming is due to human activities is not the result of a poll among climate scientists, but stems from complex statistical methods and is based on many lines of evidence, some of which are introduced below.  The fact that 97% of practicing climate scientists support the theory of AGW is actually a consequence of this extremely strong likelihood, rather than a cause, as BHS want us to believe.

BHS will rightly point out that the likelihood determination, following the IPCC guidelines, also can be done following an “elicitation of expert views”, when purely statistical methods cannot provide with a satisfying estimation. So let us review some facts BHS have forgotten but that experts know about.

Anthropogenic Warming Fingerprints

A critical step in determining the likelihood of the warming to be anthropogenic is to gather information about the phenomena.  The more data is gathered the better it is, especially if they come from distant scientific fields, because the test a given hypothesis is submitted to becomes harder to pass.  This way a hypothesis can be falsified.

There are many observations that support the theory that the recorded mean surface temperature increase is mainly caused by anthropogenic greenhouse gas emissions: less heat escaping to space, more heat radiated back to the surface at the CO2 absorption wavelengths, cooling upper atmosphere, heightened boundary between the lower and upper atmosphere, more warming during nights than days, etc.

In the meantime, the likelihood for natural factors alone to explain the increasing global temperatures and these observations is less than 10% (FAQ 9.2 of AR4WG1).

Instead of mentioning the larger temperature increases over land than oceans, which can be due to other causes than increased GHG concentrations as they rightly point out, we suggest that BHS start by scientifically refuting the above-mentioned CO2 fingerprints.  Because that would precisely falsify the CO2 hypothesis AGW theory they want so much to tear down, and perhaps the likelihood of their hypothesis would rise above 10%.  In doing so, BHS would follow the scientific method they are proud proponents of, and all awkward jumps to the conclusion that “it’s not us” will be avoided.

Cherry-picking and inventing data

A particular point BHS touch upon is the “hypothesis” that increased CO2 concentrations leads to increased water vapor content in the atmosphere.  This “hypothesis” is thermodynamically founded.  Warmer air can hold more water vapor in absolute amounts, so any factor leading to increased atmosphere temperatures ultimately leads to higher water vapor contents.  And a human-caused increased atmospheric concentration of greenhouse gases (also called heat-trapping gases) is no exception to the rules of thermodynamics, despite what BHS pretend.

Increased water vapor content and its corresponding radiative forcing have indeed been both recorded (see here for numerous references).  Dessler et al. (2008) even write that "the water-vapor feedback implied by these observations is strongly positive, with an average magnitude... similar to that simulated by climate models". The same that BHS dismiss as incapable of dealing with this issue.

Our last comment concerns their way of demonstrating how increased GHG concentrations cannot explain the recorded temperature increase. BHS rely on the GISP2 database to draw their graph. Unfortunately, GISP2 is concerned with local temperatures which do not accurately represent the average global temperature – nor even the Northern Hemisphere.  Drawing conclusions on the global climate based on GISP2 amounts to pretending that the whole world is affected by heavy rainfalls based on the precipitations in Bergen [a Norwegian city known for plentifull rainfall].

More disturbing is that as mentioned above, this critique and a rebuttal of their graph had been published by Skeptical Science some weeks before the second publication of BHS, and Prof. Humlum was contacted and stated that he would update the erroneous figure.  His way of “correcting” his mistakes does not qualify for very scientific, as is detailed in our reference.  Worse than that, comparing the simultaneous temperature and CO2 concentration is extremely misleading, as every concerned person knows that the climatic system has a large inertia.  This means that the full effects of the explosion of the CO2 atmospheric content we are causing will be seen in several decades only – but then again the GISP2 data won’t be representative of the global situation.

The observation by BHS that the mean global temperatures have dropped during the Holocene while the CO2 concentration should have led to an increase has been discussed in another post of Rob Honeycutt, and has well-known natural causes. It is wrong to infer that the current anthropogenic CO2 increase (+115 ppm from pre-industrial levels, and counting) has no effect on global warming because a 20 ppm increase in the Holocene took place as the temperature dropped slightly. The AGW theory does not pretend that natural factors cannot affect the global climate.


We have been very critical of the correspondence of BHS.  Some might even say harsh.  But we must acknowledge that their contributions are very interesting.  In fact, their letters provide with a perfect example of the strategies deployed by climate "skeptics" to twist the debate and sow doubt in the minds of the public.  BHS articulate their argumentation around the defense of an ideal of scientific method they believe in while clearly violating the rules they pretend to respect.  Citing irrelevant quotes or taken out of their context, misunderstanding fundamental concepts, concentrating on precise points without looking at the broad picture, cherry-picking or even inventing scientific facts and data in order to provide with justifications to their hypotheses, etc.

So we would like to thank them from the heart, because they have chosen to be good examples of how science should not be done, how climate science can be at worst when left to the sole hands of so-called climate "skeptics".

Notes and references:

[1] An issue with Norwegian is that there exists only one word for the two technically different concepts of probability and likelihood. The inputs to a problem define the probability of an outcome (if my dice is perfectly balanced there is an equal probability of obtaining each number from 1 to 6), whereas the likelihood of the outcome is determined as a function of the inputs (if I obtain more 6s than other numbers after a large amount of tests, the likelihood of the dice being loaded is high). Incidentally, the likelihood of an event to be due to a given factor is defined as the probability of its occurrence given the factor.

[2] Guidance Notes for Lead Authors of the IPCC Fourth Assessment Report on Addressing Uncertainties, 2004, available here. See as well IPCC Expert Meeting on Detection and Attribution Related to Anthropogenic Climate Change, 2009, available from the same website.

[3] Interestingly, the IPCC aggregates into surface warming the increasing ocean temperatures. Looking at this broader scale than merely surface atmospheric temperatures actually follows a stringent scientific approach: ignoring the increase of the ocean would amount to estimating the size of an iceberg solely by looking at its emerged tip. The informed reader will notice that so-called climate skeptics very seldom mention the ocean in their quest against the AGW theory, since doing so would severely impact their conclusions.

Posted by julienx2k2 on Monday, 23 May, 2011

Creative Commons License The Skeptical Science website by Skeptical Science is licensed under a Creative Commons Attribution 3.0 Unported License.