Greenhouse gases are responsible for warming, not the sun

June 11, 2012 | 00:00

Greenhouse gases are responsible for warming, not the sun

Scientists working on climate on a daily basis must have been rather astonished by the recent interview with professor Fritz Vahrenholt published by European Energy Review (May 2, 2012). Vahrenholt, chief of RWE Innogy, self-proclaimed climate expert and co-author of the book Die Kalte Sonne (The Cold Sun), claims that "the contribution of CO2 to global warming is being exaggerated". This claim, however, does not stand up to scientific scrutiny. We assess his ideas in the light of the scientific literature on the role of the sun versus other climate forcing factors. The dominant influence of greenhouse gases follows not only from their basic physical properties, but also from their "fingerprint" in the observed warming. The sun, in contrast, has not exhibited any warming trend over the past 50 years. The sun is thus not responsible for the warming seen during this period. Greenhouse gases in all likelihood are.

'The sun has not exhibited any warming trend over the past 50 years' (c) Thinkstock

First of all, we welcome the active participation of the business community in the discussion on climate change. Global warming and its effects may have consequences which business, e.g. the energy sector, should anticipate and adapt to. Furthermore, mitigation policies may affect the competitive advantages and business prospects of a variety of energy options. Investment portfolios should take that into account. That is not an easy task. The business consultant or director developing a climate change response strategy may be overwhelmed by the vast amount of - sometimes conflicting - scientific information available. Luckily, every couple of years an integrated assessment is made by the IPCC, the Intergovernmental Panel on Climate Change, primarily aimed at governments, but also quite valuable for the business community.

Of course, opinions differ regarding how well the IPCC assessment reflects the scientific understanding, with some claiming that IPCC overestimates the human contribution to global warming and the risks it poses, whereas others claim that these are underestimated. However, the main tenets of climate science, as described in working group 1 of the IPCC report, have proven to be robust. New research has confirmed the core conclusions, while the details are continuing to be filled in. Vahrenholt’s claim that the IPCC report is radically wrong is unfounded, and is mostly indicative of his views diverging from mainstream science.

This is not to say that no uncertainties remain; of course there are and in some cases they are inconveniently large. Inconvenient, not only because more research is required to further constrain these uncertainties, but also because the uncertainties go both ways. The human contribution to global warming could be somewhat smaller, or it could be somewhat larger than expected. Focussing only on aspects that downplay the anthropogenic contribution is closing one’s eyes to the whole picture.

Spectacular theories

Challenging the core tenets of climate science is easier said than done. Vahrenholt’s extraordinary claim requires extraordinary evidence. The least one would expect of such a claim is that it be put to the scientific test. Spectacular theories and speculations abound in books and on the internet, but most of

The human contribution to global warming could be somewhat smaller, or it could be somewhat larger than expected
them were not offered to the scientific community, or did not stand up to scrutiny if they were. Scientists routinely check each other’s work via the peer review process, which can be seen as a first test of scientific validity. From further discussions in the literature and other scientific forums the relative robustness of competing ideas is assessed. The most robust idea eventually gains acceptance. That is how science progresses. Surely the peer-review system is not perfect, but at least it is an organised process aimed at filtering what is possibly right from what is plainly wrong. Such a mechanism is lacking in the public debate; and that adds much to the public confusion about this and other complex scientific topics.

Now, would the ideas of prof. Vahrenholt stand up to scientific scrutiny? On the basis of the interview, we expect: no, they would not. However, we would still encourage him to submit his ideas for scientific review. That is where the physical forces and feedbacks in the climate system should be discussed. On the other hand, questions on how society and politics should respond cannot be answered by science, but should be discussed in the public and political debate. Unfortunately, Vahrenholt’s accusations like “we are being misinformed by the climate establishment” and “the whole purpose of the IPCC has been to get rid of the so-called Medieval Warm Period” betray him as being receptive to conspiracy theories, which are routinely echoed on the internet. In such a world view, any criticism by the scientific mainstream is of course only perceived as proof that his and similar views are being suppressed. This is often used as an excuse to not even try to submit one’s ideas to peer review. His book is criticized by scientists not because it would be politically incorrect, as Vahrenholt assumes, but because it is scientifically incorrect.

Natural fluctuations

Now, if we take a closer look at the content of Vahrenholt’s theory, his key statements appear to be the following:
1. The 'hockey stick graph' showing that global average temperatures are now higher than in the last 1500 or so years is flawed. The medieval warm period was about as warm as today, the Little Ice Age was globally very cold and both periods coincide with a more, respectively less active sun. Therefore, the IPCC underestimates the natural variations of our climate system.
2. The IPCC fails to look at all the solar characteristics, focusing on sunspots and Total Solar Irradiation (TSI) only, while both fluctuations in the magnetic field and the amplifying feedbacks should also be taken into account.
3. The sun has become more active over the past few hundred years, being responsible for a large part of the observed warming.
4. The IPCC has underestimated the natural 60 years climate cycle dominated by the Pacific Decadal Oscillation (PDO) and the Atlantic Multidecadal Oscillation (AMO).
5. Global warming has stopped since around 15 years.

On this basis, Vahrenholt claims that more warming is due to the sun and natural variability than the IPCC estimates, which automatically implies that less is due to CO2 than the IPCC says.

To begin with the last statement, the idea that global warming has stopped is a common misunderstanding, based on confusing short term variability with the long term underlying trend. Natural fluctuations, as caused by e.g. the solar cycle, volcanic eruptions, or by the El Nino/La Nina phenomenon, can mask the underlying trend in surface temperatures for more than a decade. When these effects are accounted for, as e.g. Foster and Rahmstorf did in 2011, the underlying trend is seen to continue unabated. Moreover, other measurements, e.g. of ice extent and of ocean heat content, confirm that global warming is continuing.

The solar activity has been well-measured, particularly since 1979 using satellites, and before that by indirect (“proxy”) measures. From these, the solar activity is seen to have been relatively stable over the past 50 years. That means, that even if amplified strongly, the sun's variations could still not explain the strong global warming that started halfway the 1970’s. Measurements of cosmic rays, a favourite candidate for a solar amplification mechanism, also show no trend since at least 50 years. The robust evidence needed to become a serious scientific competitor for the dominant greenhouse mechanism is sorely lacking. It is true however, that the sun gained strength over the first half of the 20th century, and thus contributed to warming seen during that time, as is also described in the IPCC report.

Various solar and climate physicists, like Lockwood, Haigh, Gray and others have published analyses indicating that the solar influence in the warming of the last half century is low or absent. These analyses include the magnetic field effects, which - in contrast to what Vahrenholt is saying - are not neglected by the IPCC. A few years ago, Pierce and Adams modeled the potential cloud forming effect of cosmic rays and found it wanting by more than an order of magnitude, even when the most favourable assumptions possible were made.

There is another indication to the statement that the sun's role in warming is limited compared to the role of greenhouse gases: fingerprints. Each possible source of warming leaves a specific and

From this quick analysis, it seems unlikely that Vahrenholt's claims would stand up to scientific scrutiny
characteristic fingerprint. For example, a solar fingerprint would be: warming throughout the atmosphere. A greenhouse gas fingerprint would be: warming at lower altitudes with simultaneous cooling of the stratosphere higher aloft, since in case of enhanced CO2 the stratosphere loses more infrared radiation than it receives from below. This is mainly a consequence of the temperature structure of the stratosphere. And guess what? Measurements clearly show a greenhouse gas fingerprint, not a solar one.

In addition, Vahrenholt claims that IPCC has underestimated natural variability, in particular a natural 60 year climate cycle manifesting in the purported Pacific Decadal Oscillation (PDO) and in the Atlantic Multidecadal Oscillation (AMO). However, Vahrenholt's statement is based on curve fitting applied to a finite time series of (local) temperatures. It is well known that curve fitting to a series of chaotic signals can lead to apparent periodicities, but that these have no value for prediction unless they are supported by an underlying physical understanding.

From this quick analysis, it seems unlikely that Vahrenholt’s claims would stand up to scientific scrutiny. They should be taken seriously, but only as an idea that deserves further research and assessment, rather than as a scientific fact or theory that rises to the level of robustness of basic climate science. In that respect, the physical science basis of the IPCC 2007 is still strong. Vahrenholt's ideas do not change that conclusion.


About the authors

Dr Rob van Dorland is a researcher at the Royal Netherlands Metereological Institute (KNMI). Dr Bart Verheggen is a researcher at the PBL Netherlands Environment Assessment Agency.

Reply to Rob van Dorland and Bart Verheggen

We thank Dr Rob van Dorland and Dr Bart Verheggen for their comments and welcome the opportunity to discuss the various points highlighted by them. First of all we have to clarify that the content of our book "Die kalte Sonne" is almost entirely based on the scientific literature. This fact is thoroughly documented in our extensive 70-pages long reference list at the end of the book. Most of the ideas we summarized in the book have already been published in peer-reviewed journals, but unfortunately they have not received the kind of attention by the IPCC that they should have deserved. Among the citations in our book are also a great number of papers that were published after the deadline of the previous IPCC 2007 report, in which Rob van Dorland served as a lead author. The suggestion by van Dorland and Verheggen to publish our "results" in a scientific journal therefore clearly does not make sense. Furthermore, rather than claiming that all these cited papers - and in fact our synthesis - are "scientifically incorrect", it would have been much better if the two scientists took the issues brought forward by us more seriously.

Van Dorland and Verheggen claim that the sun has not contributed to the warming seen during the past 50 years. Unfortunately, this claim is based on a major misconception. According to a study by Sami Solanki and colleagues that was published in 2004 in the journal Nature, solar activity during the past few decades was unusually high compared to the past 11,000 years. Van Dorland and Verheggen appear to ignore that it is not only an increasing solar trend but also solar high activity plateaus that are capable of raising global temperatures. Due to the great heat capacity of the ocean and other factors it may take several decades or longer until an external (e.g. solar) climatic impulse is fully implemented in the global climate system and equilibrium is reached (e.g. Shaviv 2005, Perry 2007, Eichler et al. 2009, Soon 2009, Wang & Zhang 2011). The high solar activity plateau of the past few decades is therefore fully compatible with the idea that solar activity has significantly contributed to global warming of the past 50 years. Reference to high frequency fluctuations on this high plateau therefore does not do the slow climate system justice. Likewise, the temperature fall from the 1940s to 1970s that coincided with a continued increase of atmospheric CO2 levels does of course not invalidate the idea that CO2 might be responsible for part of the warming experienced in the 20th century.

In fact, the best evidence for a strong climatic effect of solar activity changes comes from geological data of the past 10,000 years. In a paper published in Science in 2001 Gerard Bond and colleagues documented a series of millennial-scale climate cycles which developed in synchronicity with solar activity changes. Key drivers of these cycles are the 1000 years solar Eddy cycle and the 2300 years solar Hallstatt cycle. Since publication of this paper, these so-called Bond cycles have been found in a great number of other studies carried out in various parts of the world (see pages 68-75 in our book “Die kalte Sonne” for references). It therefore becomes ever clearer that the Bond climate cycles have to be considered a global signal, even though detailed climatic expressions depend on the respective regional settings. The now well-documented temperature development of the past 1000 years is part of this development. The Medieval Warm Period is known to have a similar temperature level as present-day and represents a solar high activity phase. The Little Ice Age has been reported from around the world and represents a phase of low solar activity.

The Hockey Stick curve once proposed by a team around Michael Mann is indeed no longer considered valid. Michael Mann and colleagues corrected it themselves in a paper they published in the Proceedings of the National Academy of Sciences in 2008. In contrast to the original hockey stick papers, the Medieval Warm Period and Little Ice Age in the 2008 publication suddenly re-appeared. This new, more correct version was essentially confirmed by independent work by Fredrik Ljungqvist published in 2010.

The warming of the past 300 years clearly follows the natural pattern we have experienced over the past 10,000 years in the geological history. While the sun has to be considered to be a major driver of this development, also anthropogenic CO2 must have played a role. One of the key tasks of today’s climate sciences should be to quantitatively determine the roles of natural and anthropogenic climate divers. Current climate models used by the IPCC only assume a very small effect of the sun. With this assumption, however, the models are not able to reproduce the significant temperature and climate roller coaster during pre-industrial, post-glacial times. It is not plausible that the sun should have suddenly stopped influencing climate significantly over the past 50 years when it has done so for the entire past 10,000 years. Contrary to what Van Dorland and Verheggen claim, it is not just “details” in the climate models that need to be adjusted, but rather faulty core assumptions that lead to a failure in climate hindcasts of the past 10,000 years. And models that are not able to reproduce the past 10,000 years are of course unlikely to be able to predict the future.

Empirical evidence from the past 10,000 years clearly shows that some sort of a solar amplifier mechanism has to exist. Several amplifier candidates have been proposed and published in the peer-reviewed literature. These include for example mechanisms via the solar magnetic field, cosmic rays and clouds proposed by Henrik Svensmark as well as an amplifier effect based on UV radiation effects in the stratosphere. Research is still ongoing in this important field. While some papers fail to find an effect, others do indeed prove important elements of these proposed amplifier mechanisms. Cherry picking of negative papers does not do the important scientific discussion justice.

Unfortunately the IPCC has chosen not to implement any of these solar amplifier candidates in their models, despite the clear evidence for the existence of such processes from the historical postglacial data. At the same time, a highly questionable, very powerful water vapor/cloud amplifier is implemented in the models that multiplies the climate effect of CO2 up to several times.

Van Dorland and Verheggen also claim that we proposed that global warming has stopped. The contrary is the case. Apparently they have not had the chance to read our book. In fact we schematically predict an additional warming until 2100 by up to 1°C. In the coming decades, however, ocean cycles and low solar activity may cause a mild cooling by a few tenths of a degree, before temperatures pick up again due to the CO2 greenhouse effect. Furthermore it is a well-established fact, that temperatures have not increased since at least the year 2000, which however should not be misunderstood that warming on a 30 year scale might have stopped.

Van Dorland and Verheggen also incorrectly assume that the 60 year ocean cycles would represent mainly chaotic noise. Apparently they are not aware of key geological papers in which the Atlantic Multidecadal Oscillation (AMO) and the Pacific Decadal Oscillation (PDO) have been reconstructed back for the last 1000 years and longer (e.g. publications by Glen MacDonald and Roslyn Case in the Geophysical Research Letters in 2005 and Mads Knudsen and colleagues in Nature Communications in 2011). These papers have impressively documented the existence of a clearly detectable cycle with a period ranging between 50 and 80 years. The cycles are empirically known to have significantly modulated the longer term temperature trends by a few tenths of a degree which needs to be taken into account in the climate models. Interestingly, lack of warming since 2000 could have been readily predicted based on ocean cycle patterns, and was in fact forecast by Mojib Latif who recently said for this reason that warming will not significantly pick up for another 10 years.

Finally, Van Dorland and Verheggen mention an alleged stratospheric fingerprint that was originally thought to prove the significant climate effect of CO2. A look at the data, however, reveals that temperature in the stratosphere has essentially remained stable since 1995. The observed

From this quick analysis of Van Dorland and Verheggen's comments, it is evident that the current models proposed by the IPCC suffer from a number of major weaknesses which deserve urgent attention
stratospheric cooling was restricted to the phase 1980-1995 and notably coincided with the decrease of the ozone concentration in the ozone layer, most probably related to anthropogenic CFC-emissions. A recent modeling paper published by Berger and Lübken in the Geophysical Research Letters has now confirmed that the cooling in the middle atmosphere is largely due to this ozone effect. Stratospheric temperature trends therefore can no longer be considered as valid proof for a significant climatic effect of CO2.

From this quick analysis of Van Dorland and Verheggen's comments, it is evident that the current models proposed by the IPCC suffer from a number of major weaknesses which deserve urgent attention. Contrary to the impression which the two scientists try to convey to the readers, the physical science basis of the IPCC 2007 models is far from being robust and convincing. We urge that our criticism be taken seriously. Suitable independent re-evaluations may need to be initiated because the implications of faulty climate models would be very serious in light of the costly political decisions currently being prepared on their basis.


Short reply to Fritz Vahrenholt and Sebastian Lüning

In their reply to our criticisms, Prof Fritz Vahrenholt and Dr Sebastian Lüning exhibit a misunderstanding of various key aspects of climate science. Their claim, that mainstream science is radically wrong, is unfounded and not supported by sufficiently strong evidence. We will focus here on a few major issues of contention.

  • Vahrenholt and Lüning are correct in pointing out that the climate has a delayed response to changes in its radiation budget (whether from the sun, from CO2 or from other causes) due to the ocean’s heat capacity. This may cause the planet to continue to warm after the radiative forcing has stabilized, but the rate of warming will decrease and level off as the climate equilibrates to the new situation.
  • However, the observations show that both surface temperatures as well as ocean heat content started to increase (during the 1970's and 80's) long after solar activity had reached its plateau (during the 1950's). This is inconsistent with a lagged response to the sun, as suggested by Vahrenholt and Lüning. The relatively steady rate of warming of both ocean and atmosphere over the past four decades indicates that this must be caused by another process. The sun cannot be responsible for this particular warming, irrespective of how strongly one wishes to amplify its effect.

Updated graphic of total heat content from Church et al 2011 (via

Vahrenholt and Lüning cite the work of Solanki and co-authors in support of their claim. However, Solanki et al made the same point as we do: "This comparison shows without requiring any recourse to modeling that since roughly 1970 the solar influence on climate (through the channels considered here) cannot have been dominant" (Solanki et al., 2003), and: "Although the rarity of the current episode of high average sunspot numbers may indicate that the Sun has contributed to the unusual climate change during the twentieth century, we point out that solar variability is unlikely to have been the dominant cause of the strong warming during the past three decades." (Solanki et al., 2004). 

  • Vahrenholt and Lüning correctly conclude that fluctuations in short term trends "should not be misunderstood that warming on a 30 year scale might have stopped". In contrast, in the original EER interview Vahrenholt said “Our critics say fifteen years is not enough to make judgments about the climate.” Apparently, he now agrees with his critics.
  • Vahrenholt and Lüning claim that stratospheric temperatures have remained stable, even though they are expected to decrease in response to a stronger greenhouse effect. Again they conclude something about a long term trend based on short term fluctuations, against their own advice. Moreover, the paper by Berger and Lübken which they cite is irrelevant because it presents temperature trends in the mesosphere, an even higher altitude. The CO2 fingerprint is mainly confined to the stratosphere and is distinct from the effect of ozone. The latter also induces cooling, but mainly above the poles, while increasing CO2 cools the stratosphere everywhere. The point remains that if the sun were the dominant cause of surface warming, this warming would also have occurred at higher stratospheric layers. This is not what has been observed. 
  • Vahrenholt and Lüning confuse what are assumptions versus what are emergent properties or outcomes of climate models. For example, the amplifying feedback by water vapor is a result of the interplay of various elements of atmospheric physics as incorporated by General Circulation Models. They also state that the small effect of solar variability is an assumption of climate models, whereas it is an outcome.

Climate changes in the deep past (going back hundreds, thousands or even millions of years) cannot be explained, let alone quantitatively modeled without a substantial warming effect from CO2.

Wishing away the effects of CO2 is not enough in science; it needs to be quantitatively demonstrated
We invite readers to view this excellent talk by the American geologist Richard Alley. It's telling that no physics-based climate model has been developed that can simulate past and recent climate changes without a substantial warming effect from an increase in greenhouse gas concentrations. Wishing away the effects of CO2 is not enough in science; it needs to be quantitatively demonstrated.

Our purpose is to advance understanding of all factors responsible for the recent anomalous increase in global temperatures. We encourage readers to peruse the hyperlinks in our first reply for additional background information. Not all the additional points brought up by Vahrenholt and Lüning in their reply are relevant to their central thesis. However, these points will be discussed at Bart Verheggen's blog at


Loading comments...
related items