Responses to the Climate Science Survey

Appeared in similar form on the PBL website

In the Spring of 2012, the Netherlands Environmental Assessment Agency PBL held a survey among 1868 scientists studying various aspects of climate change, including physical climate, climate impacts, and mitigation. The main results of the survey were published in an article in Environmental Science and Technology (ES&T) in August 2014: “Scientists’ views about attribution of global warming”. It showed that there is widespread agreement regarding a dominant influence of anthropogenic greenhouse gases on recent global warming. This agreement is stronger among respondents with more peer-reviewed publications.

A background report with the results for all 31 questions has now been made available. The total number of responses for each answer option is provided and a subdivision into seven groups for five questions. The background report contains previously unpublished data. Some highlights are provided below.

Climate sensitivity

Respondents were asked for their opinion regarding the best estimate and likely range for equilibrium climate sensitivity (ECS). This is an important quantity for projections of global warming, as it gives the expected warming that would follow from a doubling in atmospheric CO₂ concentration after the climate system has equilibrated to the new conditions. Thus, expected warming in the future depends on the combination of total emissions and climate sensitivity.

The figure below gives the average estimates of ECS from 7 groups of respondents, including authors of the Working Group I report of the fourth IPCC Assesment Report (AR4), respondents who signed public declarations critical of mainstream climate science as embodied by IPCC (‘unconvinced’), and four different subgroups distinguished according to their self-declared number of climate related peer-reviewed publications (0–3; 4–10; 11–30; more than 30). Results from most groups were very close to the IPCC range (1.5-4.5 ^°C) mentioned in the fifth assessment report (AR5) – except those tagged as ‘unconvinced’ which strongly deviated from the other groups, and to a lesser extent the group of respondents with three or less publications. For all subgroups the ‘best estimate’ was slightly lower than the ‘best estimate’ reported in AR4 (i.e. 3 ^°C). AR5 provided no best estimate.

Role of climate science in society

Respondents were also asked their opinion about seven statements regarding the role of climate science in society and how the science should be communicated. There was a strong consensus that scientists themselves should communicate with both policymakers and the general public about climate change and that communication with the general public should focus on solid knowledge. To a lesser extent there was agreement that risks and uncertainties should be emphasised during such communication. Responses varied more strongly about whether or not existing uncertainties in climate science strengthen the case for mitigation (i.e. to avoid potential low probability, high impact events). There was (strong) disagreement with the statement that climate science would be too uncertain to be useful for policymaking on climate change.

The role of the sun in global warming

In the public debate about climate change the role of the sun is often put forward as an alternative explanation for global warming. Question 17 asked what fraction of recent global warming could be attributed to the sun. Those tagged as ‘unconvinced’ had the lowest fraction of respondents that indicated that they don’t know (together with AR4 authors) and the highest fraction that said that the role of the sun is unknown. As expected they also had by far the highest fraction (27%) that believed that the sun caused more than half of recent global warming.

As with the attribution questions (see the ES&T article), there appears to be a trend in responses going from the group with fewest publications to those with most. The more publications about climate change respondents report to have written, the larger fraction of them agree with the IPCC position that the sun hardly played a role in recent global warming, since the solar output decreased slightly over that period.

More information:

• To the background report
• ES&T article (Open Access)
• Comment on Scientists’ Views about Attribution of Global Warming
• Reply to Comment on Scientists’ Views about Attribution of Global Warming
• Blogpost with the ES&T article
• Video-interview with Collin Maessen about the ES&T article
• FAQ’s with the ES&T article

PS: I’ll have a poster presentation about the survey at the EGU conference this week, in session EOS6 “Communication and Education in Geoscience” on Thursday evening.

Advertisement

ClimateDialogue on Climate Sensitivity

After a bit of a “hiatus”, ClimateDialogue (CD) has re-opened again with a discussion on climate sensitivity. On the one hand this site is unique in bringing together ‘mainstreamers’ and ‘contrarians’ (both in the organization and in the discussions), hopefully leading to both enhanced clarity on what the (dis)agreements are really about and to decreased polarization. On the other hand it’s controversial because a ‘false balance’ is embedded in its structure (by purposefully inviting contrarian scientists to the discussion, rather than e.g. randomly inviting experts).

Whether the positives or negatives dominate is in the eye of the beholder (opinions about that vary wildly), but also depends very strongly on the participation of the mainstream (both as invited experts and as contributing to the public discussion). See also my initial reflections at the time of the first launch. Discussions on ClimateDialogue will be facilitated and moderated by Bart Strengers (NL Environmental Assessment Agency, PBL) and Marcel Crok (freelance journalist), where the former has a mainstream view of climate science and the latter a contrarian view. I am still involved in the background, as is KNMI (NL Meteorological Institute). ClimateDialogue is funded by the Dutch Ministry of Infrastructure and Environment.

In the current ‘dialogue’ James Annan, John Fasullo and Nic Lewis are discussing their views about climate sensitivity (the equilibrium warming after a doubling of CO2 concentrations, ECS). In the latest IPCC report (AR5) the different and partly independent lines of evidence are combined to conclude that ECS is likely in the range 1.5°C to 4.5°C with high confidence. The figure below shows the ranges and best estimates of ECS in AR5 based on different types of studies, namely:

– the observed or instrumental surface, ocean and/or atmospheric temperature trends since pre-industrial time

– short-term perturbations of the energy balance such as those caused by volcanic eruptions, included under “instrumental” in the figure

– climatological constraints by comparing patterns of mean climate and variability in models to observations

– ECS as emergent property of global climate models

– temperature fluctuations as reconstructed from palaeoclimate archives

– studies that combine two or more lines of evidence

Eric Wolff on areas of agreement and on the public debate about climate science

Dr. Eric Wolff is spot on (see also further below): 

as an outsider to the blogosphere, it surprises me that so many people, presumably mostly with even less knowledge and training than me, seem absolutely convinced they have mastered every area of climate science.

A peculiar line-up of speakers assembled recently at the Conference on the Science and Economics of Climate Change in Cambridge: Phil Jones, Andrew Watson, John Mitchell, Michael Lockwood, Henrik Svensmark, Nils-Axel Morner, Ian Plimer, Vaclav Klaus and Nigel Lawson. Bishop Hill reports that

Dr Eric Wolff of the British Antarctic Survey tried valiantly to find a measure of agreement between the two sides.

This proved an interesting exercise and resulted in a useful list (also reproduced by Judith Curry) on what we can all agree on (perhaps excluding those too far out on the fringes; links added by me):

1. CO2 does absorb infrared radiation
2. The greenhouse effect (however badly named) does occur in practice: our planet and the others with an atmosphere are warmer than they would be because of the presence of water vapour and CO2.
3. The greenhouse effect does not saturate with increasing CO2
4. The CO2 concentration in the atmosphere has risen significantly over the last 200 years
5. This is because of anthropogenic emissions (fossil fuels, cement production, forest clearance). [Ian Plimer disagreed, but as Curry stated: “the anthropogenic contribution is (should be) undisputed”]
6. If we agree all these statements above, we must expect at least some warming. [Bishop Hill noted “broad agreement” with this point]
7. The climate has warmed over the last 50 years, [as is evident from] land atmospheric temperature, marine atmospheric temperature, sea surface temperature, and (from Prof Svensmark) ocean heat content, all with a rising trend.
8. We probably don’t agree on what has caused the warming up to now, but it seemed that Prof Lockwood and Svensmark actually agreed it was not due to solar changes, because although they disagreed on how much of the variability in the climate records is solar, they both showed solar records without a rising trend in the late 20th century. [Excellent point and ironically hitting Svensmark with a stick of his own making. Note the distinction between variability and trend.]
9. On sea level, I said that I had a problem in the context of the day, because this was the first time I had ever been in a room where someone had claimed (as Prof Morner did) that sea level has not been rising in recent decades at all.  I therefore can’t claim we agreed, only that this was a very unusual room.  However, I suggested that we can agree that, IF it warms, sea level will rise, since ice definitely melts on warming, and the density of seawater definitely drops as you warm it.
10. Finally we come to where the real uncertainties between scientists lie, about the strength of the feedbacks on warming induced by CO2 [i.e. climate sensitivity]

Eric Wolff also chimed in with a substantial comment over at Bishop Hill’s, rebutting some commonly heard arguments and making some very spot-on remarks:

(…)

I should first state the rationale for the summary I made at the Downing event. The meeting was about the science and economics of climate change, and I was asked to lead a discussion that came between the science talks and the two economics talks. I therefore felt the most useful thing I could do was to try to summarise what we had heard, as a basis for the discussion of whether society should do anything in response to that, and if so what. In particular I did hear a surprising number of things on which almost everyone in the room could agree, and it seemed worth emphasising that, rather than rehearsing old arguments.

I notice in the thread here several comments about who sets the “terms of the debate”, and about the “context of the debate”. While such phrasings may make sense in discussing energy policy, it is a strange way to discuss the science. Our context is the laws of physics and our observations of the Earth in action; our aim as scientists is to find out how the Earth works: this is not a matter of debate but of evidence. I think some of the comments on this blog come dangerously close to suggesting that we should first decide our energy policy, and then tell the Earth how to behave in response to it.

Regarding Plimer’s proposal that volcanic emissions were more important than we thought, (…) if volcanoes were “causing” the recent increase, then around 1800 their emissions would have had to rise above their stable long-term rate, and then stayed high. This is however a rather hypothetical discussion because the change in the isotopic composition of CO2 in the atmosphere over the industrial period is not consistent with an increased volcanic source anyway.

Regarding the idea that the “temperature increase stopped in 2000”: my point is that we know there are natural variations (due eg to El Nino) that cause runs of a few years of temperature colder than the average or a few years warmer. Look on the record at the decade around 1910 for example when there was a long run of cold years on a flat background.

Such a period superimposed on a trend would look like the last decade (but more so). The point of my analogy is that you can’t determine the trend over several months by measuring the gradient in a run of a few days. Similarly, you simply can’t determine a multidecadal trend by measuring the gradient over a few years because all you get is the “noise” of natural variability.

Professor Morner claims that globally sea level has not risen at all; he dismisses the evidence, from both satellites and the global tide gauge network, that it has. (…) There are numerous reasons why a single site can show a sea level signal, either real or apparent, that departs from the global mean.

I am not the best person to discuss models and feedbacks in detail (see comment on expertise below). However, I could not let two issues pass. Firstly, when models are run out for a century into the future, they do indeed show runs of years with flat temperatures amidst a trend (…) (because of El Nino and other natural factors). I am therefore not clear why this is evidence that something is missing. Regarding positive feedbacks: a positive feedback implies amplification, but not a system out of control; this is only the case if the sum of the gain factors is greater than 1.

Finally, a few specific issues that interested or worried me.

(…) CO2 emits infrared as well as absorbing it. (…) indeed, this property is precisely why its effects do not saturate (but fall logarithmically), because it allows the height from which the emitted radiation finally escapes to rise into regions with less and less air.

Geckko accidentally made an important point. S/he did not like the statement: “We can agree that if it warms the sea level will rise”, because it was too simplistic. Well, as a scientist I always like to boil things down to a statement that my brain can grasp, but that contains the essential explanation of an observation or process. And this one does, for example being demonstrably what was observed in going from a cold ice age world, with sea level 120 metres below the present level, to the present. However, you are right: there are factors that could make this statement false, such as increased snowfall when it warms, adding more ice into ice sheets. As soon as several such competing processes have to be taken into account, our brains cannot predict the outcome, and so we have to resort to putting all the “millions of assumptions” into a numerical model and seeing which of them “win”. An argument for models?

Coldish made a good point about expertise, and this is where I am going to go into a slightly more challenging area. I freely admit that I am not an expert on all, or even most, aspects of climate. When I reach a topic that I have not previously studied, I go to those who are experts, either in person or by reading their work. I maintain scepticism about some of their conclusions, but my working assumption is that they are intelligent and that they have probably thought of most of the issues that I will come up with. Can I observe as an outsider to the blogosphere, that it surprises me that so many people, presumably mostly with even less knowledge and training than me, seem absolutely convinced they have mastered every area of climate science.

and more so, convinced that they are right and almost all of the experts are wrong. That must be the height of hubris.

However, Coldish specifically mentioned IPCC, and I think there is also an interesting point to make about that. At the Downing event, there seemed to be two IPCCs in the room. To some it was a huge plot, masterminded by some mysterious power that manipulates troublesome scientists. To me and the scientists in the room, it is (at least in WG1) simply a set of well-researched review papers, describing the present state of the peer-reviewed literature. I mention this only because I think the former view is a type of groupthink where, because people form an extreme opinion in their private space, they think it is widely held, or even true.

Alas, Wolff’s forray into the blogosphere was short, as is evident from a short comment a while later:

Just in case anyone thinks they are addressing me with their remarks:

I thought this might indeed be a chance for a civilised discussion, and some of the respondents seem happy to have that. However there are also a lot of remarks on here that are frankly rude and aggressive, and I won’t be returning. Now I remember why I hate the blogosphere.

Radiative forcing by aerosol used as a wild card: NIPCC vs Lindzen

(This is also featured as a guest post on Skeptical Science)

The greatest source of uncertainty in understanding climate change is arguably due to the role of aerosols and clouds. This uncertainty offers fertile ground for contrarians to imply that future global warming will be much less than commonly thought. However, some (e.g. Lindzen) do so by claiming that aerosol forcing is overestimated, while others (e.g. the NIPCC) by claiming that aerosol forcing is underestimated. Even so, they still arrive at the same conclusion…

Let’s have a look at their respective arguments. Below is a figure showing the radiative forcing from greenhouse gases and from aerosols, as compared to pre-industrial times. The solid red curve gives the net forcing from these two factors. The wide range of possible values is primarily due to the uncertainty in aerosol forcing (blue dotted line).

The greenhouse gas forcing (dashed red curve) is relatively well known, but the aerosol forcing (dashed blue curve) is not. The resulting net anthropogenic forcing (red solid curve) is not well constrained. The height of the curve gives the relative probability of the associated value, i.e. the net climate forcing is probably between 1 and 2 W/m², but could be anywhere between 0 and 3 W/m². (From IPCC, 2007, Fig 2.20)

NIPCC’s argument

The NIPCC report (a skeptical document, edited by Craig Idso and Fred Singer, made to resemble the IPCC report) says:

“The IPCC dramatically underestimates the total cooling effect of aerosols.”

They hypothesize that natural emissions of aerosol precursors will increase in a warming climate, causing a negative feedback so as to dampen the warming. Examples of such gaseous aerosol precursors are dimethyl sulfide (DMS) emitted by plankton or iodocompounds created by marine algae. They use these putative negative feedbacks to claim that

“model-derived sensitivity is too large and feedbacks in the climate system reduce it to values that are an order of magnitude smaller.”

Rebuttal

These are intriguing processes (the “CLAW hypothesis” first got me interested in aerosols, when I assisted with DMS measurements on some remote Scottish islands), but their significance on a global scale is ambiguous and highly uncertain. As a review article about the DMS-climate link says:

“Determining the strength and even the direction, positive or negative, of the feedbacks in the CLAW hypothesis has proved one of the most challenging aspects of research into the role of the sulfur cycle on climate modification.”

The NIPCC report exaggerates the uncertainty in climate science, but seems to put a lot of faith in elusive and hardly quantified processes such as natural aerosol feedbacks coming to our rescue.

Lindzen’s argument

On to Lindzen:

“The greenhouse forcing from man made greenhouse gases is already about 86% of what one expects from a doubling of CO2 (…) which implies that we should already have seen much more warming than we have seen thus far (…).”

Lindzen again (E&E 2007):

“How then, can it be claimed that models are replicating the observed warming? Two matters are invoked.”

The “two matters” he refers to are aerosol cooling and thermal inertia from the oceans (a.k.a. “warming in the pipeline”). He then proceeds to argue that both of these factors are much smaller than generally thought, perhaps even zero. E.g. on aerosols, Lindzen writes:

“a recent paper by Ramanathan et al (2007) suggests that the warming effect of aerosols may dominate – implying that the sign of the aerosol effect is in question.”

By downplaying the importance of these two factors, Lindzen argues that the observed warming implies a small climate sensitivity. The same line of argument is also used by Dutch journalist Marcel Crok, who writes in his recent book (in Dutch, my translation):

aerosols probably cool much less than commonly thought

Rebuttal

While it is true that aerosols can warm and cool the climate (by absorption and reflection of solar radiation, respectively, besides influencing cloud properties), most evidence suggests that globally, cooling is dominant. Whereas Ramanthan et al (2007) don’t quantify the net aerosol effect (in contrast to Lindzen’s implicit claim), Ramanathan and Carmichael (2008) (quoted by Crok) do. They estimate both the warming ánd cooling effects to be stronger than most other estimates, but the net forcing (-1.4 W/m²) is right in line with (even a little stronger than) the IPCC estimate (see the above figure). Taking into account realistic estimates of aerosol forcing and ocean thermal inertia, the earth has warmed as much as expected, within the admittedly rather large uncertainties. Ironically, it is exactly because aerosol forcing is so uncertain and because the climate hasn’t equilibrated yet that the observed warming since pre-industrial times is only a very weak constraint on climate sensitivity. Lindzen seems very certain of something that most scientists would readily admit is very uncertain.

Conclusion

So we have the peculiar situation that both of these approaches try to claim that climate sensitivity is small, but the NIPCC approach is to claim that aerosol forcing is very large (thus providing a negative feedback to warming), whereas the Lindzen approach is to claim that aerosol forcing is very small (thus necessitating a small sensitivity to explain the observed warming so far). Of course they can’t both be right, and probably neither of them are. Looking back at the figure above, both approaches are based on assuming that aerosol forcing is at the edge of the probability spectrum (as if it were some fudge factor), whereas the most likely value is somewhere in the mid range. Both approaches also ignore the other lines of evidence that point to climate sensitivity likely being in the range of 2 to 4.5 degrees. E.g. a value as small as suggested by the NIPCC (0.3 degrees) is entirely inconsistent with the paleo-climate record of substantial climate changes in the earth’ history. And finally, both approaches implicitly assign high confidence to some of the most uncertain aspects of climate science, even though they routinely mock climate science as if nothing is known at all.

Of course it is not mandatory for all those who dismiss mainstream climate science to agree, but to see two important “spokespeople” for climate contrarians take such mutually inconsistent approaches is peculiar. Even more so when you realize that Lindzen signed the recent “prudent path” letter to US Congress, in which the NIPCC report was approvingly cited… Most people can’t have it both ways, but apparently climate contrarians can.

See also a more detailed critique of the NIPCC and of Lindzen’s argument.