The 97% consensus of climate scientists is only 47%

Summary: In February 2014 I examined surveys of climate scientists and found (as had others) that they showed broad agreement with the IPCC’s headline statement about warming since 1950. However time brings new research, such as a major survey that digs deeper and finds that only a minority of climate scientists agree with the full key statement of AR5 about greenhouse gases — the most recent IPCC report. That’s important news.  Also see the important update below.

Update: fame from Politifact!

The good liberals at Politifact did a hit pieced on this post. Tskilled disinformation with the assistance of climate warriors in academia. It’s an interesting story of noble lie corruption, which I describe in

This post produced quite a frenzy among the alarmists. Linda Qiu Politifact published a bizarre rebuttal, ignoring what I said and replying to things I didn’t say (this is a favorite tactic of alarmists). She recruited professors to do so, because science! For details see Politifact tells us about American politics and science. We should listen.

The Survey: finding the consensus

In March – April 2012 the PBL Netherlands Climate Assessment Agency, with several other scientists, conducted a survey of approximately 6,550 scientists studying climate change. It was published as “Scientists’ Views about Attribution of Global Warming” by Bart Verheggen et al in the 19 Aug 2014 issue of Environmental Science and Technology (peer-reviewed). In April 2015 they published a more detailed report (used in this post).

This survey covered many of the frontiers of climate science. This post examines one the questions about the keynote statement of the IPCC’s most recent work at time of the study — Assessment Report 4 (AR4, published in 2007). {This is a correction from the original post, which looked at the headline statement of AR5, about all forcings}. From AR4’s Summary for Policy-makers:

“Most of the observed increase is global average temperature since the mid-20th century is very likely due to the observed increase in anthropogenic greenhouse gas concentrations.”

In 2013 the IPCC published AR5, which repeated this finding  — but on page 884, in Chapter 10 of WGI:  “We conclude, consistent with Hegerl et al. (2007b) {i.e., chapter 9 of AR4}, that more than half of the observed increase in GMST {global mean surface temperature} from 1951 to 2010 is very likely due to the observed anthropogenic increase in GHG {greenhouse gas} concentrations.”

The PBL survey is the first I’ve seen to test agreement with both facets of these statements. First, how much of the global surface warming is caused by anthropogenic (human-caused) emissions of greenhouse gases? (Note AR5 referred to all factors; see “Details” below). Only 1,222 of 1,868 (64% of respondents) agreed with AR5 that the answer was over 50%. If we exclude the 164 (8.8%) “I don’t know” respondents, 72% agree with the IPCC. So far, so good.

PBL survey: question 1a

Now for the second part of the statement: what is the certainty of this finding? That the IPCC gives these answers is one of its great strengths. Of the 1,222 respondents to the PBL survey who said that the anthropogenic contribution was over 50%, 797 (65%) said it was 95%+ certain (which the IPCC defines as “virtually certain” or “extremely likely”).

PBL survey: question 1b

Those 797 respondents are 43% of all 1,868 respondents (47% excluding the “don’t know” group). The PBL survey finds that only a minority (a large minority) of climate scientists agree with the AR4 keynote statement {and the similar finding in AR5’s chapter 10} at the 95% level typically required for science and public policy {Note: the last section added for greater clarity}.

Update: reconciling the PBL survey results with AR4 & AR5

Tom Curtis (attorney) posted a comment at Skeptical Science, that put the PBL survey results in the proper context of AR4 and AR5. Kudos to him for this excellent work!

From AR4’s Summary for Policy-makers: “Most of the observed increase is global average temperature since the mid-20th century is very likely due to the observed increase in anthropogenic greenhouse gas concentrations.” Published in 2007, this reflected the consensus at that time.

In 2013 the IPCC published AR5, which repeated this finding  — but on page 884, in Chapter 10 of WGI:  “We conclude, consistent with Hegerl et al. (2007b) {i.e., chapter 9 of AR4}, that more than half of the observed increase in GMST {global mean surface temperature} from 1951 to 2010 is very likely due to the observed anthropogenic increase in GHG {greenhouse gas} concentrations.”

GHG being, of course the focus of proposed public policy changes to mitigate climate change.

AR5 shifted the headline SPM finding to “extremely likely” about “all anthropogenic forcings”. This was widely but mistakenly reported as an increase in their confidence level about anthropogenic warming. Even some (many?) climate scientists believed that the IPCC had increased its confidence level about anthropogenic forcings from AR4 to AR5 (e.g., this interiew with Prof Judith Curry).

Stand back I'm trying science.

Conclusions

Scientists, like experts of all kinds, often say they “just know” things for which there is uncertain or contradictory research. A massive body of research shows that such opinions are often wrong. That’s why we rely on the power of science to give more reliable answers, and on organizations like the IPCC to help us understand the current state of knowledge about climate change. The IPCC is a political entity, but it is the best we have.

But the challenge of climate change — and the trillions it will cost to mitigate — require a clear view of what’s known, with what degree of certainty. But instead we’ve been told increasingly fanciful tales of what “97% of climate scientists” believe, often things far beyond the most confident statements in the IPCC’s AR5.

This latest survey suggests that even the IPCC might not represent the consensus as accurately as previous surveys research indicated. Only 64% of climate scientists agreed that over half of the warming since 1950 was from anthropogenic factors, and only 65% of those had a confidence level of 95%+ — so that only 43% agree with the full keynote statement of AR5. That’s important by itself, and tells us much about the accuracy of what we read in the news media about climate science.

Many scientists have warned us of this problem.

“The drive to reduce scientific uncertainty in support of precautionary and optimal decision making strategies regarding CO2 mitigation has arguably resulted in unwarranted high confidence in assessments of climate change attribution, sensitivity and projections…”

— “Climate Science and the Uncertainty Monster” by Judith Curry (Prof Atmospheric Science, GA Inst Tech), April 2015.

Also: Is there a minority viewpoint or theory in climate science?

The PBL survey shows substantial minority viewpoints in the many specific questions they examine. The study does not look for patterns, to discover if there is one or more “skeptic” theories opposing the consensus — the dominant paradigm, to use Thomas Kuhn’s The Structure of Scientific Revolutions. Criticism sometimes improves theories, but institutions often ignore critics. Either way the critics will likely have only minor effects; the climate’s evolution will resolve these debates.

Kuhn’s work shows that a paradigm cannot be disproved, only replaced (details here). Unless the skeptics form a theory, they’ll remain minor players in the climate science debate. The public policy debate about climate change (they’re distinct, although often conflated) operates more like a court: the defense (take no action) wins if they raise a reasonable doubt about the threat of anthropogenic climate change (this was revised in response to a comment).

Note that preparing for occurrence of past extreme weather would help prepare us, and that should command support from both sides. I suspect it will continue to get support from neither side because politics rules.

For more about the two debates see this by political scientist Don Atkin: “There are two debates about climate change“.

Updates

(1)  There are always messy details, so these numbers have to be interpreted broadly. In AR4 the  keynote finding was that over half of the warming was caused by greenhouse gases — with 90% certainty. AR5 said the certainty was 95% for all anthropogenic forcings. This relevant question in the PBL survey referred only to greenhouse gases. The difference gets into some complex matters, explained in the EST paper under “Aerosol Cooling Versus GHG Warming”.

(2)  A report confirming the findings reported here: “97 consensus? No! Global warming math, myths, and social proofs” by the Friends of Science, 17 February 2014. Thoroughly documented; 48 pages long.

(3)  I suspect that the PBL study is a clear case of “p-hacking”, slowly emerging to view as a major problem in modern science. See this from the excellent “Psychology is in crisis” article at VOX.

Take Joseph Hilgard, a psychologist at the University of Pennsylvania. In a recent blog post titled “I was wrong,” he fesses up to adding a shoddy conclusion to the psychological literature (with the help of colleagues) while he was a graduate student at the University of Missouri. “[W]e ran a study, and the study told us nothing was going on,” he writes. “We shook the data a bit more until something slightly more newsworthy fell out of it.”

This a bold and honest move — the type that gives me reasons to be optimistic for the future of the science. He’s confessing to a practice called p-hacking, or the cherry-picking of data after an experiment is run in order to find a significant, publishable result. While this has been commonplace in psychology, researchers are now reckoning with the fact that p-hacks greatly increase the chances that their journals are filled with false positives. It’s p-hacks like the one Hilgard and his colleagues used that gave weight to a theory called ego depletion, the very foundation of which is now being called into question.

The authors of the PBL survey were looking to confirm the results of the AR4 headline statement about GHG (relegated to the footnotes in AR5, replaced by one about  “all anthropogenic forcings”). They found that only 43% of all 1,868 respondents agreed at the 95% confidence level. They buried this result and shifted the emphasis to who agreed with it. Unfortunately their survey was not designed to elicit robust information about this. Exactly as described in the VOX article.

But there was a problem: The experiment found no effects for game violence or for game difficulty. “So what did we do?” Hilgard writes. “We needed some kind of effect to publish, so we reported an exploratory analysis, finding a moderated-mediation model that sounded plausible enough.” They found that if they ran the numbers accounting for a player’s experience level with video games, they could achieve a significant result. This newfound correlation was weak, the data was messy, and it barely touched the threshold of significance. But it was publishable. …

There are a few big problems with this. The biggest is that their experiment was not designed to study experience level as a main effect. If it had, they perhaps would have done a better job of recruiting participants with varying ranges of experience. “Only 25 people out of the sample of 238 met the post-hoc definition of ‘experienced player,'” Hilgard writes. Such a small sample leaves the study with much less statistical power to find a real result.

Postcards from the frontier of science

For More Information

If you liked this post, like us on Facebook and follow us on Twitter. For more information see The keys to understanding climate change and My posts about climate change. Also, see these posts about the IPCC…

  1. “Climate Change: what do we know about the IPCC?”, 2
  2. Climate scientists speak to us. What is their consensus opinion?
  3. The IPCC releases its advice on “Impacts, Adaptation and Vulnerability”. To be attacked from both sides.
  4. Another disturbing article about climate change. Fortunately we have the IPCC!
  5. Is our certain fate a coal-burning climate apocalypse? No!

To help you better understand today’s extreme weather

To learn more about the state of climate change see The Rightful Place of Science: Disasters and Climate Change by Roger Pielke Jr. (Prof of Environmental Studies at U of CO-Boulder, and Director of their Center for Science and Technology Policy Research).

154 thoughts on “The 97% consensus of climate scientists is only 47%

  1. Pingback: Big Survey Finds Only Minority Of Climate Scientists Agree With IPCC Keynote Statement | The Global Warming Policy Forum (GWPF)

  2. mtobis (@mtobis)

    Polls of this sort depend on the population interviewed.

    I am not dissuaded by the poll from my very strong perception that at most hardly anybody who understands the physics of the situation has more than the tiniest residual doubt that human influences are dominating contemporary climate change.

    The survey was designed to be maximally inclusive of “climate scientists” rather than focusing on people informed about climate physics itself. Additionally, it bends over backwards to give so-called skeptics a voice. The measured consensus that emerges on physical questions is something of a lower bound, not a balanced estimate of genuinely expert opinion.

    The IPCC process is designed to provide a conservative estimate of the opinions of the people with the most relevant expertise on a given question. Among those expert on the question at hand that you have decided to focus on, I continue to maintain that there is very little serious dispute.

    The likelihood that the net anthropogenic forcing component post 1950 is 100% imposed on a continuation of the background natural cooling of the past several millennia, a possibility that notoriously confuses Dr. Curry.

    People who have been following this question professionally are very unlikely to disagree because the evidence is indeed very strong. A consensus document is intended to weigh evidence, not to take a poll on consensus on individual questions. Specific assertions are authored by the community with the relevant expertise.

    It would be interesting to see how responses to this question shake out among those respondents willing to respond to the question about the Charney sensitivity.

    Liked by 1 person

    Reply
    1. Editor of the Fabius Maximus website Post author

      Michael,

      Thank you for providing additional color on this issue!

      As for your disagreements with the data shown here, the data is what it is. It’s what we have until better data comes in. Analysis of survey data (like all data) should avoid interpretation by “just so stories” that express confirmation bias. Experts are just as prone to this as anyone, but often have unjustified confidence in their immunity from it. This is a commonplace in political debates, which are often ruled by epistemic closure (I listen only to those who agree with me; others are prima facie invalid.)

      As for minority opinions among climate scientists, the survey did not explore patterns of belief among participants. However, the data suggests that there is a 10-20% minority cohort.

      “Polls of this sort depend on the population interviewed.”

      I agree, which is why the authors made such a strong effort to get a representative sample. A team including John Cook can hardly be accused of bias towards skeptics.

      “on the question at hand that you have decided to focus on”

      I assume you meant that as humor. The public policy discussion about climate change has been dominated for 20 years about the “consensus” of climate scientists. First, represented by the IPCC. In the last few years by the claims of a “97%” consensus. If you believe this focus is inappropriate the time to protest was when you began posting on the web in 2007.

      Background for readers

      from Michael Tobis’ profile at Planet3.0:

      e is editor-in-chief of Planet3.0 and site cofounder, has always been interested in the interface between science and public policy. He holds a doctorate from the University of Wisconsin – Madison in Atmospheric and Oceanic Sciences where he developed a 3-D ocean model on a custom computing platform. He has been involved in sustainability conversations on the internet since 1992, has been a web software developer since 2000, and has been posting sustainability articles on the web since 2007.

      Liked by 2 people

    2. Fernando

      I wouldn’t have used that 95 % sure in view of the medieval warm and the unusual temperature swings I’ve seen postulated from isotope records. The climate seems to be loaded with feedbacks, as far as I know they aren’t fully explained and the models can’t capture the system dynamics.

      You know, sometimes this 95 % reminds me of the way Texas Aggies are so certain their team will win really big. They tend to give too many points and lose their bets when they play Oklahoma and Texas.

      Liked by 1 person

  3. socialbill

    As a number of people have remarked science is not democratic.

    Before 1998 the consensus in the astronomical community was that the expansion of the universe was either slowing somewhat or asymptotically toward zero. Preliminary data seemed to indicate one of these two outcomes. If a vote had been taken only about 5÷ (not including me) would have voted for acceleration, the true answer. The New Hubble based data on supernovae distances were finally good enough to completely overturn this consensus. This was because:

    1) this situation is simple compared to climate science and
    2) there weren’t trillions of $ riding on the outcome.

    Perhaps the solution is simply better data.

    Liked by 1 person

    Reply
    1. Editor of the Fabius Maximus website Post author

      Social Bill,

      This is an example of what I noted in this post, conflating the science debate with the public policy debate. The “consensus” — more usefully, the paradigm — is vital in both.

      To say that consensus (the paradigm) doesn’t matter in science is typical normative nonsense. Science is conducted neither by machines or Vulcans, but by people for whom social dynamics trump any pretty principles. The teams, especially when there is a dominant paradigm, matter much for little things like getting published, getting a job, getting promoted, getting awards and such.

      In the public policy debate about science matters the size of the consensus and its confidence level are paramount factors. Both described by the IPCC for each of its major findings (it was, after all, designed as a political machine to help decisionmakers with public policy questions). There is no other factor remotely so useful to evaluate such matters.

      Like

  4. Evert Wesker

    Basically the first question is: Are human activities (greenhouse gas emissions due to these) causing a part of the current warming of the climate. As far as I’m concerned the answer on this question is a resounding “yes”.

    But then comes the second question: How large is this part exactly? And then (some) uncertainly comes into play. If one looks at the “climate sensitivity” numbers (the expected temperature rise due to a doubling of the CO2 equivalent greenhouse gases concentration in the atmosphere) there is a considerable margin of uncertainty. See as an illustration from various sources.

    Add to this the various “wildcards” (ocean oscillations, solar activity and volcanism) and the “absolute” certainty is to some extent a bit more subtle.

    So is this “97% consensus questioning” giving us fundamental new insights? Not as far as I’m concerned. IPCC AR5 still stands. However, one should not read it selectively (and jump to “simple” conclusions) either.

    Finally: The precautionary principle is about posterity. We are not the last humans on this planet – at least, that’s my approach. But that holds for many subjects (quality of living, energy, environment, natural resources, climate). It is all about striving for sustainability in general.

    Mazzel & broge, Evert Wesker

    Like

    Reply
    1. Editor of the Fabius Maximus website Post author

      Evert,

      “the “climate sensitivity” numbers (the expected temperature rise due to a doubling of the CO2 equivalent greenhouse gases concentration in the atmosphere) there is a considerable margin of uncertainty.”

      That’s quite an understatement. The range of scientists’ estimates of ECS in this survey is large, and that’s for only a 66% confidence level (1 standard deviation). They don’t show the 2 SD range, but it must be so enormous to be of little use for public policy work. Especially since the estimated range has been decreasing during the past decade. I don’t understand your point.

      “the “absolute” certainty is to some extent a bit more subtle.”

      I don’t know what that means. There is no such thing as “absolute certainty” in these matters, or in any significant public policy debate.

      “So is this “97% consensus questioning” giving us fundamental new insights? Not as far as I’m concerned.”

      It appears that’s because you have not understood the point of the post.

      “The precautionary principle is about posterity.”

      Yes, it’s “for the children” — the standard emotional appeal made when lacking data to justify the sought public policy action. It’s a daft basis for action. For details see The first step to protecting the world from its many dangers.

      Like

    2. Ali Bertarian

      Regarding:
      “(1) I believe the precautionary principle is not operationally useful for public policy. For detail see The first step to protecting the world from its many dangers.

      (2) There is nearly zero evidence that an ice age is likely during the 21st century, the maximum time horizon for public policy.”

      The claim that public policy is limited to the 21st century, even if true, could mean that the precautionary principle is just being ignored due to limitations of government effectiveness; it does not mean that another Ice Age in the 22nd, 23rd, or subsequent centuries is not a greater risk factor to human life.

      Check the two world maps of temperature data for the year 2014 provided by NOAA. There are no real polar data, most of Africa has no real data, and much of Asia has no real data. Yet the map on the right disingenuously claims that we have real temperature data from which future temperatures can be calculated. Estimated or interpolated data are not real data for the calculation of risk. Is that how you want the FDA to calculate risk when they test experimental drugs? We aren’t discussing science anymore; we are discussing subjective feelings of fear.

      Another assumption made by pro-AGW people is that there are only risks involved with higher temperatures, not benefits. Was human life better off in the Medieval Warm Period or during the Little Ice Age?

      Liked by 2 people

    3. Editor of the Fabius Maximus website Post author

      ALi,

      “it does not mean that another Ice Age in the 22nd, 23rd, or subsequent centuries is not a greater risk factor to human life.”

      I suggest you read my post about assessing risks. That is exactly the sort of shockwave that can be assessed only by comparison with the many other risks we face.

      Like

  5. thomaswfuller2

    Thanks for posting on this survey. I’ve been writing about it for a couple of years. You’ve put up a good and interesting post.

    I think you go a bit far with your inference regarding 47%. 66% by my count agreed that half or more of recent warming was caused by human emissions. Those who agreed were far more certain of their conclusions than those who felt human contributions were lower.

    The question was not phrased ‘Do you agree with the IPCC statement that it is 95% certain… etc.” I think you go a bit too far with your statement. This is because you are reporting on responses to two different questions.

    However, it is clearly not 97%. I have written Verheggen about some of his statements regarding the survey results–he sort of goes too far in the other direction, IMO.

    As I told him, there is a protocol for reporting the results of this type of survey. You start with “X% indicated that half or more of recent warming is due to….”

    Then you show the data for each response. Then you report on subgroups. Then you address the second question. Then you start to make inferences. ‘It could be argued that the actual percentage of those agreeing with the IPCC was 47%.” But without the preparation, your statement will attract criticism that perhaps it does not deserve.

    Like

    Reply
    1. Editor of the Fabius Maximus website Post author

      Thomas,

      “You go too far”

      I don’t understand your objection. They made the comparison to the AR5 statement. The survey provided the subgroups (results by answer to the question and follow-up question). I just showed the math.

      “It could be argued”

      The results of their questions were clear. No inference required. I don’t believe you read the survey accurately.

      Liked by 1 person

    2. Editor of the Fabius Maximus website Post author

      Thomas,

      I seldom give much attention to these “scientists don’t know basics of science” comments. But this morning I realized my reply might have been incorrect. I said you didn’t carefully read the survey. But you probably didn’t carefully read the post.

      “Then you show the data for each response. Then you report on subgroups. Then you address the second question. … But without the preparation….”

      Both the post and the report clearly state that the authors did exactly what you described. Question 1a gives the initial response. 1b gives the relevant subgroup (those who agree with both parts of the AR5 WG1 keynote statement). Divide the two by the total number of respondents gives the fraction agreeing with the proposition being tested.

      Why would you believe that a team of scientists — and their expert reviewers — would incorrectly do something so basic?

      Like

  6. Pingback: What consensus? Less than half of climate scientists agree with the IPCC “95%” certainty | The Authentic Male

  7. Gail Combs

    I noted your first commenter said “People who have been following this question professionally are very unlikely to disagree because the evidence is indeed very strong.”

    Well, not exactly. The following new information is why you are not going to get a ‘Consensus’. I was fortunate enough to see a graduate level physics lecture on the subject that was very well presented and was a presentation of actual real world data. “Why has there been no global warming for the past decade?” by William Happer (Prof Physics, Princeton), 8 Sept 2014. Slides 16, 22, 42, 43 and 44 are the critical slides.

    A less-technical lecture for the lay person: {file is corrupted; will not show}

    Dr Robert Brown, a physicist at Duke University also commented on the subject {comment posted at Watts Up With That, 28 May 2014}.

    The take away from this lecture (9/2014) is CO2 ‘modeling’ is a mish-mash of theoretical equations and experimentally derived data. Where the Climate modelers missed the boat is in using equations for ‘line broadening’ aka the ‘wings’ where the current CO2 absorption ( at 400 ppm) is supposedly taking place. These equations produce results that do not match up to the experimental data. The lines are not as broad as theory would have it. SEE Slide 22: Lorentzian line shape nor Voigt line shapes are correct in the far wings! This was the point of the lecture.

    Since the experimental data shows less broadening this flattens the logarithmic curve and essentially lowers the ‘Climate Sensitivity’ of CO2 for a doubling from 400 ppm to 800 ppm to less than 1C===> 0C {Graph}

    The other point is very slow partial radiative decay rates of CO2 molecules were determined by experiment. At the surface the time it takes for an excited CO2 model to radiate is ten times longer than the time between collisions so at the surface the energy is handed off via collision. At the surface the CO2 is also saturated which is why the line broading is so important and why the logarithmic curve is so flattened after 250 ppm. What was found is the curve is even flatter than the theoretical curve. Dr. Brown also makes the same point.

    What does this mean? This means the energy CO2 absorbs from the earth is handed off to non green house gases increasing their temperature and adding to convection. Also answered in the lecture was my question about where CO2 energy is radiated instead of being handed off via collision. Experimental data shows barely any radiation at 11 KM and that CO2 is radiating in the stratosphere ~ 47 KM above the surface. you can see that in this graph.

    The fact that CO2 is radiating above the troposphere makes the next point critical too.

    The overlap with the pure-rotational band of water vapor eliminates most of the response from the lower band edge, and IR from clouds further reduces the response to more CO2. The true response is likely less than half the ideal limit or a doubling is less than 3.7Wm^2″

    You can get useful background for understanding the physics from WIKI. SUBJECTS:
    Mössbauer effect (recoil energy lost during absorption <===CRITICAL)
    Motional narrowing
    Voigt effect
    The Pound–Rebka experiment (VERY IMPORTANT because gases are moving randomly and in random directions)

    …The test is based on the following principle: When an atom transits from an excited state to a base state, it emits a photon with a specific frequency and energy. When an atom of the same species in its base state encounters a photon with that same frequency and energy, it will absorb that photon and transit to the excited state. If the photon’s frequency and energy is different by even a little, the atom cannot absorb it (this is the basis of quantum theory). When the photon travels through a gravitational field, its frequency and therefore its energy will change due to the gravitational redshift. As a result, the receiving atom cannot absorb it. But if the emitting atom moves with just the right speed relative to the receiving atom the resulting doppler shift cancels out the gravitational shift and the receiving atom can absorb the photon….

    Liked by 2 people

    Reply
  8. Pingback: The Climate Change Debate Thread - Page 5011

  9. Pingback: The 97% consensus of climate scientists is only 47% | The Morning Show with Toni and Gary

  10. Robert Clark

    Another big takeaway from this is that even among climate scientists there still is not overwhelming agreement that most of global warming is human caused. A large percentage are still undecided or go the other way that it’s mostly natural.
    So if you had 5 climate scientists together at random and someone came over inviting them to a rally about AGW, 3 might say “Yeah, I’ll be there”, but 2 would say “Yeah, go knock yourself out”.
    Hardly, overwhelming support.

    Bob Clark

    Liked by 1 person

    Reply
    1. Editor of the Fabius Maximus website Post author

      Robert,

      I agree, and was surprised that only 64% expressly agreed that over half of the warming since 1950 was anthropogenic. The 10% “unknown” and 9% “don’t know” answers were astonishing. The Environmental Science and Technology article amazingly ignored the major findings of their study, which I suspect is why my post 18 months later has gotten so much attention.

      The response from some climate scientists on Twitter has been interesting. First, that the sample is unrepresentative of climate scientists — perhaps, but any study including John Cook (of Skeptical Science) as an author was obviously designed to support the activists’ case.

      Second, that the competent climate scientists agreed most strongly with the IPCC. Which might be so, but they didn’t use a strong measure of competency (e.g., an H index). Number of publications is pretty crude, and doesn’t reflect if the scientist is an academic or practitioner.

      Third, and most amusing, that the IPCC”S AR5 Working Group I authors most strongly supported the keynote statement of the IPCC’s AR5 WG1. Yes, but that’s a measure of institutional affiliation — not competency.

      Liked by 1 person

  11. Robert Clark

    On another forum someone noted another odd aspect of this survey, that 17% said the human caused contribution was above 100%. He suggested we should therefore discount that group altogether. Then less than 50% would say most global warming is human caused.
    Any explanation in the paper why they even put that in as an option?

    Bob Clark

    Liked by 1 person

    Reply
    1. Marco

      Because there is data that indicates the combined non-anthropogenic forcings are negative and without anthropogenic forcings the earth would thus be expected to be cooling. cf http://www.skepticalscience.com/pics/KnuttiAttributionGraph.png

      There is also the issue of greenhouse gas emissions and attribution, where anthropogenic aerosols are removed from the equation. In that case the attribution is well over 100%.

      There thus is nothing odd with that option.

      Like

    2. Editor of the Fabius Maximus website Post author

      Marco,

      “Because there is data that indicates the combined non-anthropogenic forcings are negative”

      Perhaps so, but the graph you cite does not show that. It show a few brief dips below zero. The graph is from “Anthropogenic and natural warming inferred from changes in Earth’s energy balance” by Markus Huber & Reto Knutti, Nature Geoscience, Jnauary 2012.

      “without anthropogenic forcings the earth would thus be expected to be cooling. … There thus is nothing odd with that option.”

      Here are the 4 emission scenarios evaluated in the IPCC’s latest report, AR5. The green line shows CO2 emissions if we institute drastic policy changes to mitigate climate change: they continue until after 2070. I know of nobody who considers this likely to take begin in the next decade (who can say after that?) — so any natural cooling is a distant event.

      So there is nothing “odd” about a scenario of 21st C cooling; it’s just extremely unlikely.

      IPC's AR5 WG1: 4 scenarios (RCPs) of emmissions

      “There is also the issue of greenhouse gas emissions and attribution, where anthropogenic aerosols are removed from the equation. In that case the attribution is well over 100%.”

      Yes, aerosols are cooling agents. Why is that an “issue”?

      Like

    3. Editor of the Fabius Maximus website Post author

      Robert,

      “another odd aspect of this survey, that 17% said the human caused contribution was above 100%.”

      That’s a complex issue; the person on the forum does not understand the issue. The survey asks about the “human caused contribution” of warming from greenhouse gases (GHGs). Many climate scientists believe that aerosols cool the climate, so the effect for GHGs is larger than for total anthro forcings.

      AR5 states that since 1950 the total anthro forcing is equal to the warming. Some believe that for some period (not necessarily since 1950) the anthropogenic forcing is greater than 100% of observed warming — i.e., that the natural trend is cooling. See this graph from “Anthropogenic and natural warming inferred from changes in Earth’s energy balance” by Markus Huber & Reto Knutti, Nature Geoscience, January 2012 (click to enlarge):

      Huber-Knutti Attribution Graph

      Like

    4. Marco

      The graph I cite *does* show that. Take 1950 to present (in that graph) and the trend is (very slightly) negative.

      In other words, in the absence of the anthropogenic forcings, based on that specific study (there are other similar studies) we would have expected the earth’s atmosphere to cool (very, very slightly) since 1950, and thus anthropogenic forcings will have caused more than 100% of the warming. Thereby the “odd aspect” that Bob Clark thinks he has identified is actually not an “odd aspect” at all.

      I pointed out the anthropogenic aerosols because survey questions are not always understood the way they were conceived to be understood, and thus some of those who filled out the survey may have solely considered (consciously or subconsciously) the impact of anthropogenic greenhouse gas emissions in this specific question. In such a case the attribution is well over 100%.

      Like

  12. Science of Doom

    The papers reviewed for that chapter of the IPCC report basically assume the models correctly model internal variability. Some don’t mention it. Some mention it in passing. This unproven assumption is the basis of the statistical certainty.

    The authors of chapter 11 of the same IPCC report pointed this out and, because the models are well-known to under-state structural uncertainty, gave a “pragmatic” update of the certainty to a completely different value. The updated “pragmatic” value has zero statistical credibility but is clearly more reliable than the 95% certainty.

    Explained in Natural Variability and Chaos – Seven – Attribution & Fingerprints Or Shadows?

    Like

    Reply
  13. donaitkin

    Quintus,

    You may be right in saying that unless skeptics can put together a theory they will remain minor players in the science debate about climate. But I don’t think that is the case with respect to the policy debate. There the issues are, at least in principle, much clearer. The MAGICC calculator shows that no matter how much we reduce greenhouse gas emissions, global temperature will stay largely unaffected. Why therefore are we doing this? The policy outcomes involve costs and benefits, and are easier to argue on the part of those who don’t come from the lab bench.
    Otherwise, I agree with your post and its conclusions.

    Liked by 1 person

    Reply
    1. Editor of the Fabius Maximus website Post author

      donaitkin,

      You raise an important point — the need for a competing theory is much greater in science than in the public policy debate. Politics is more like law, in that only a reasonable doubt need be raised. I’ll made a correction.

      Thanks for catching this poor logic!

      Like

  14. Pingback: 95%, attribution, and all that! | …and Then There's Physics

  15. Joshua

    Personally, I ALWAYS go with what Richard Tol has to say.

    “Published papers that seek to test what caused the climate change over the last century and half, almost unanimously find that humans played a dominant role.”

    I love how for some “skeptics,” referring to the prevalence of agreement among climate scientists is an appeal to authority and antithetical to “true science,” …er…until it isn’t.

    Like

    Reply
    1. Editor of the Fabius Maximus website Post author

      Joshua,

      Re: appeal to authority

      That seems a bit binary, and probably a false dilemma.

      I do not believe any discussion can occur about science or public policy without some appeal to authority. Nobody has all knowledge, let alone the ability to research everything from first principles.

      As with most things in life, getting useful answers requires using tools — like appeal to authority — at the right time, in the right way. It is not simple or easy, hence the debates.

      Like

  16. Joshua

    Fabius –

    I agree. The concept of (falacious) appeal to authority is tribalistically exploited by both sides in the climate wars, as are the relevance/importance/magnitude of agreement among experts on climate science.

    Liked by 1 person

    Reply
    1. Editor of the Fabius Maximus website Post author

      Joshua,

      I think the role and agreement of experts gets confused because both sides (or all sides) conflate the science and public policy debates. They interrelate but run by different “rules.”

      Consensus has little or no normative role in science, but plays a powerful role in the social dynamics of science. That was Thomas Kuhn’s big insight (i.e., role of paradigms), and has been developed by decades of research since then.

      For public policy, determining the degree and breath of the consensus of scientists is often a key input. Hence the value of the IPCC giving a confidence level for each of its major findings — and of surveys testing those (“trust but verify”).

      Like

  17. Joshua

    I don’t accept the little or no role in a scence (noting that you qualified or with normative) because I don’t accept notion of pure science that stands apart from the social role of science. Imo, they are inextricably linked as no scientist stands apart from the society in which they do their science.

    We routinely accept the existence of a predominance of view among experts on complex matters that are beyond our pay grade, as being informative if not dispositive. Consider the broad acceptance of a scientific consensus on whether HIV causes aids. Hardly anyone argues that a consensus on that issue has no normative role in the science. The fact that the minority views are a minority is a relevant part of the science. Of course researchers consider the predominant interpretation of evidence as relevant. It doesn’t mean that they necessarily accept the predominant view…but any scientist would necessarily consider the prevalence of interpretation of complex evidence as they evaluate the validity of their own interpretations.

    But either way, that is a side issue that is interesting philosohically but it’s a red herring w.r.t.the climate wars, imo. what I clearly see are partisans who seek to exploit the notion of “pure” science to advance their social/policy agenda. Their interpretation is the pure one, and those who disagree are allowing their view of science to be biased.

    Like

    Reply
    1. Editor of the Fabius Maximus website Post author

      D blake,

      “The Science isn’t settled”

      I agree. Oddly enough, claims that a scientist said “the science is settled” were an urban legend — until July when Marcia McNutt — editor-in-Chief of Science and next President of the National Academy of Sciences — said “The time for debate has ended. Action is urgently needed.“ Details here.

      Like

  18. Pingback: Verheggen’s Consensus: Not 97%, not 47%. It’s 66%. | The Lukewarmer's Way

  19. thomaswfuller2

    I take issue with your analysis here: “Verheggen’s Consensus: Not 97%, not 47%. It’s 66%“.

    “Fabius Maximus writes, “Now for the second part of the statement: what is the certainty of this finding? That the IPCC gives these answers is one of its great strengths. Of the 1,222 respondents to the PBL survey who said that the anthropogenic contribution was over 50%, 797 (65%) said it was 95%+ certain (which the IPCC defines as “virtually certain” or “extremely likely”).”

    But the IPCC offers a definition of 95% certainty in their publications. This definition was not presented to survey respondents prior to asking them about how certain they were. Some may have used the IPCC definition of certainty, some may not. In any case, without knowing how they defined certainty it is not possible to assert that only 47% of respondents agree with the consensus.

    We do know that 66% agree. We know that they are more confident than those who disagree. That is the percentage that should be offered as the Consensus. It is exactly the same percentage as found in a survey by Bray, von Storch et al in 2008 (although the question was worded differently), which should give us more confidence in the figure.”

    For what it’s worth I don’t agree with Verheggen’s reporting either.

    Liked by 1 person

    Reply
    1. Editor of the Fabius Maximus website Post author

      Thomas,

      “Some may have used the IPCC definition of certainty, some may not.”

      That’s a valid point. Considering the fantastically high profile of the IPCC during the working career of these climate scientists — and the extreme importance of confidence limits — I wonder how many were not aware of this. I’m in finance, write about geopolitics, and I’ve been aware of the definition of 95% for many years. It looks to me like a minor point. How much different could there be in people’s definition of “virtually certain”?

      It certainly seems minor compared to the larger questions — such as the population of respondents, and the views of respondents and non-respondents. Every survey has limitations. The PBL survey looks to me like a large improvement over previous surveys (as should be expected). Hopefully future work will address the questions raised by the PBL survey.

      “that should be offered as the Consensus.”

      I disagree. Confidence limits are an essential input to the public policy process. Over-simplification is a source of errors and confusion. There have been far too much of both in this debate.

      However, there is no way to resolve this different of opinion — and my experience with this post provides additional confirmation that this debate is polarized — indeed polluted — beyond hope. Climate science as an institution has become dysfunctional, and only weather will resolve this debate. My follow-up post will discuss this.

      Like

  20. thomaswfuller2

    Hi Fabius Maximus,

    Just to continue briefly, I agree that confidence limits are important to energy policy discussions. But just because they are important doesn’t mean we can impute them where they are in fact not defined in the data.

    It is precisely because confidence limits are important that we should be very conservative at implying their existence where we in fact do not know what they are.

    It’s a limitation of the survey that we only realize post facto due to the analysis choices made by the report authors. That happens a lot in surveys. I strongly disagree with the emphasis of the report on number of publications as a determining factor for the consensus. I think it’s absurd that they do not report the top line figures for the questions involved.

    We however can be confident of the 66% agreement figure and I think honest discussion of the survey’s implications should stick to that.

    Liked by 1 person

    Reply
    1. Editor of the Fabius Maximus website Post author

      Thomas,

      We can agree to disagree, but pointlessly because this is correct:

      “The time for debate has ended”
      — Marcia McNutt (editor-in-Chief of Science, next President of the NAS) in “The beyond-two-degree inferno“, Science, 3 July 2015.

      The public policy debate in the US is effectively over. Not only will we not take substantive effects to prepare for possible future climate change, we will not prepare for reoccurrence of past weather. Climate science as an institution is broken, the larger science community applauds its dysfunctionality, and a critical mass of the US public has lost confidence in it. The weather will determine future policy.

      The public policy debate is much like those that dominate military affairs today. Which battleship was the best? With equal arms could a Roman legion defeat Henry V’s army at Agincourt? it’s fantasy football. Will world temp in 2100 be 2C higher than preindustrial or a thousand? Whatever.

      Much as the climate science debate moved from “is there a pause” to “causes of the pause” to “forecasting duration of the pause” — the public policy debate should move to determining Lessons Learned from this debacle. That’s the subject of my next post.

      Liked by 1 person

  21. thomaswfuller2

    I have frequently compared the blogosphere discussion to fantasy football, usually adding that we are privileged that sometimes the football players actually show up in the corner bar where we’re fighting and add their two cents’ worth.

    I don’t think the debate is over. I actually don’t think the proper terms of the debate have been set. Fools from the Konsensus side have been saying the debate is over and the science is settled for more than a decade. But the debate continues.

    Your assessment of the current state of affairs can also be simply translated as ‘six years of a Democratic administration’ and is subject to change at the next election. I am a Democrat and will most likely vote for the Democratic candidate in 2016–but I don’t take that or the current state of affairs for granted.

    Liked by 1 person

    Reply
    1. Editor of the Fabius Maximus website Post author

      Thomas,

      Time will tell who is correct. However, note I’m good at forecasting (it’s my business). See my record.

      “Your assessment of the current state of affairs can also be simply translated as ‘six years of a Democratic administration’ and is subject to change at the next election.”

      Nothing is happening. Do you believe a GOP administration will do more than Team Obama? Also, I’ve seen no forecasts that the House or Senate will change hands in 2016, so even Hillary is unlikely to be able to do much (if she tries, which I doubt).

      Like

  22. ...and Then There's Physics

    I’ve just realised something. I think you’ve fallen foul of the Prosecutor’s Fallacy. The attribution study that lead to the suggestion that it’s extremely likely that most of the warming since 1950 was anthropogenic, was actually a hypothesis test. The attribution studies attempt to show that the warming since 1950 could be explained in the absence of anthropogenic forcings and find that this is only possible in a small percentage of cases (95% sure that it was anthropogenic, because we haven’t actually tested this explicitly, we’ve simply rejected the hypothesis that it could be explained in the absence of anthropogenic forcings. Hence, you’ve essentially addressed this the wrong way around. James Annan discusses this briefly here.

    Like

    Reply
    1. ...and Then There's Physics

      Somehow, I managed to completely mess up the above comment. This is an attempt to remember the bits that I somehow managed to delete

      The attribution studies attempt to show that the warming since 1950 could be explained in the absence of anthropogenic forcings and find that this is only possible in a small percentage of cases (95% sure that it was anthropogenic, because we haven’t actually tested this explicitly, we’ve simply rejected the hypothesis that it could be explained in the absence of anthropogenic forcings.

      Like

    2. Editor of the Fabius Maximus website Post author

      ATTP,

      You’ve not explained why the mechanics of attribution studies are relevant to this discussion. You appear to consider attribution studies as some form of faith-based revelation. Which is fine. As for the rest of us — scientists have some degree of confidence in their output. This will increase or decrease over time as additional information becomes available.

      Like

    3. ...and Then There's Physics

      Blast, it’s done it again. Doesn’t like greater than and less than signs. Okay, third go.

      The attribution studies attempt to show that the warming since 1950 could be explained in the absence of anthropogenic forcings and find that this is only possible in a small percentage of cases (less than 5%) and so they reject this hypothesis and accept that most of the warming since 1950 was anthropogenic. Strictly speaking we can’t say that we’re more than 95% sure that it was anthropogenic, because we haven’t actually tested this explicitly, we’ve simply rejected the hypothesis that it could be explained in the absence of anthropogenic forcings.

      Like

    4. ...and Then There's Physics

      Really? So, you’re incapable of actually having a discussion about this and you feel sorry for me. I guess being condescending is all you have? Anyway, I suspected this was going to be a waste of time, and it seems I was right.

      Like

    5. Editor of the Fabius Maximus website Post author

      ATTP,

      We were never having a dialog because you cannot state my proposition, and reply instead to incorrect versions. That’s fact. I repeatedly explain this, which you ignore. That’s sad. What more is there to say?

      You could try to state my proposition, and list your objections. That would provide a basis for discussion.

      Like

    6. ...and Then There's Physics

      You appear to consider attribution studies as some form of faith-based revelation.

      So, I point out that attribution studies are a scientific analysis and you conclude that I think they’re a faith-based revelation. Given that you clearly do not want to discuss this, I’ll simply point out this strange assessment and stop.

      Like

    7. Editor of the Fabius Maximus website Post author

      ATTP,

      I stated that because you appear to believe it inappropriate to evaluate scientists’ degree of confidence in attribution studies. When I point this out, you discuss their mechanics — as if that makes them beyond question. It doesn’t.

      Like

    8. Editor of the Fabius Maximus website Post author

      ATTP.

      Paraphrase of post: 797 respondents said that >50% of warming since mid-20th can be attributed to GHG AND have a confidence level of “extremely likely” or “virtually certain” (which AR4 & AR5 define as >95%). That is 43% of all 1,868 respondents (47% excluding the “don’t know” group). Those combined statements are an aprox equivalent to the keynote statements in AR4 (and close to that in AR5). This is the best survey testing climate scientists’ consensus on this point. The answers to question 1a is in close agreement with previous surveys. The responses to 1b add a valuable new dimension to this, an important input to the public policy debate.

      Like

    9. ...and Then There's Physics

      When I point this out, you discuss their mechanics — as if that makes them beyond question

      I’ll try to explain again why the mechanics are relevant. The “extremely likely” in the attribution study actually refers to the level of confidence at which we reject that more than 50% of the warming could be non-anthropogenic, not to how sure we are that it is anthropogenic. Not only have you analysed the survey participants confidence in their OWN level of anthropogenic warming (which isn’t simply “more, or less, than 50%”) it’s also not consistent with the IPCC’s attribution statement which – more correctly – is the confidence we have in rejecting the hypothesis that more than 50% is non-anthropogenic.

      Like

    10. Editor of the Fabius Maximus website Post author

      ATTP,

      We’ve gone over this before. When looking at surveys you inject your own beliefs on the data. The PBL study is quite clear about the limitations of these studies’ ability to show what people mean by these responses, and the danger of over-interpreting them.

      Otherwise we get rebuttal by stories. You believe the respondents had an identical precise definition of theses confidence levels. Another guy believes that climate scientists interpretation of “extremely likely” (precisely defined in AR4 & AR5) varies so much that these results are useless. Perhaps you’re both right, or wrong. Social science surveys are only probes into complex matters, each revealing incremental information — and providing pointers to areas needing more research.

      Survey data has to be used carefully or it becomes fodder for motivated reasoning. We see this in the belief that earlier studies with less precise data were definitive, whereas the incremental data in this should be discarded. Perhaps so. I’ll wait for further research to tell us.

      Like

    11. Editor of the Fabius Maximus website Post author

      ATTP,

      Correction! I said “We’ve gone over this before.” That was wrong. I had this discussion with M Tobis on Twitter about a survey discussing the pause. Still, this illustrates the problem with guessing about meaning of questions. He quickly runs through 3 radically different definitions of the pause, all while remaining confident that his definition is correct.

      Me: “Are people confused by question’s lack of time parameter: year, decade, century, millennia? How would a scientist answer?”
      MT: “it’s in the present tense. So the answer is unambiguously and (for practical purposes) certainly yes.”
      AT: “I don’t understand. Warming “in the present tense” means over what time horizon? Makes no sense if present=instantaneous.”
      MT: en.wikipedia.org/wiki/Derivative
      Me: “You believe a public opinion poll asks for the derivative of global surface temperature, rather than average change over a period?”
      MT: “Q seems to mean “is it warmer than it would have been otherwise & going to get warmer”? Time period not implicit or needed.”
      Me: “You have an interpretation; others can have different meanings. Big literature about question design. This Q is poorly written.”

      Like

    12. ...and Then There's Physics

      Fabius,

      Paraphrase of post: 797 respondents said that >50% of warming since mid-20th can be attributed to GHG AND have a confidence level of “extremely likely” or “virtually certain” (which AR4 & AR5 define as >95%). That is 43% of all 1,868 respondents (47% excluding the “don’t know” group).

      Which is what I understood it to be. Read the statement again. It doesn’t say “What confidence level would you ascribe to more than 50% of the warming since 1950 being anthropogenic?”, it says “What confidence level would you ascribe to YOUR ESTIMATE ….”. Since the levels of warming were narrower than simply “more than 50%” and “less than 50%” these are not equivalent. Also, there’s a vast difference between asking someone what confidence interval the scientific anaylsis suggests, and what confidence they have personally.

      Furthermore, as I’ve tried to explain a number of times, the confidence level in the attribution statement actually refers to the level of confidence at which we reject the hypothesis that more than 50% could be non-anthropogenic, not to the level of confidence associated with accepting that more than 50% could be anthropogenic. Not only is your analysis not consistent with the consensus position, it also seems to be a double test – testing both what fraction accept that most of the warming since 1950 is anthropogenic AND what fraction regard there own chosen warming as extremely likely/virtually certain.

      Like

    13. Editor of the Fabius Maximus website Post author

      ATTP:

      “Since the levels of warming were narrower than simply “more than 50%” and “less than 50%” these are not equivalent.”

      Why?

      “there’s a vast difference between asking someone what confidence interval the scientific anaylsis suggests, and what confidence they have personally.”

      I believe that’s a speculative interpretation of question one: “what can be attributed to”. The question is poorly written (esp the phrase “can be”), probably because they didn’t consult someone with relevant experience in survey design.

      “as I’ve tried to explain”

      As I’ve said, you state that without any supporting evidence, and in defiance of the study’s explicit warning about over-interpretation of results due to different interpretations of the questions. For example, if 1a referred to GHG alone or all anthro forcings.

      Like

    14. ...and Then There's Physics

      Social science surveys are only probes into complex matters, each revealing incremental information — and providing pointers to areas needing more research.

      It’s also worth doing a sanity check. An analysis that concludes that the level of consensus (however defined) is only 47% does not pass it.

      Like

    15. ...and Then There's Physics

      It’s difficult to take such statements seriously, or have any interest in a dialog with someone who considers such statements a response.

      It’s similarly difficult to take seriously, ot have any interest in dialog with, someone who thinks that the level of consensus is around 47%.

      Seriously, this might be snarky, but 47% is really absurd. It’s one thing to want to be somewhat contrary. It’s another to take up a position that is hard to take seriously.

      Like

    16. ...and Then There's Physics

      Why?

      Look at question 1a. It wasn’t simply “more than 50%” and “less than 50%”.

      I believe that’s a speculative interpretation of question one: “what can be attributed to”. The question is poorly written (esp the phrase “can be”), probably because they didn’t consult someone with relevant experience in survey design.

      That doesn’t invalidate my point.

      As I’ve said, you state that without any supporting evidence, and in defiance of the study’s explicit warning about over-interpretation of results due to different interpretations of the questions. For example, if 1a referred to GHG alone or all anthro forcings.

      I linked to James Annan’s post which links to this. In the latter post, read the bit starting “The real probabilistic meaning of the 95% figure”. It’s also pretty standard null hypothesis testing. The null hypothesis would be that more than 50% of the warming since 1950 is non-anthropogenic (or less than 50% is anthropogenic). If there is a less than 5% chance of this, then you reject the null at the 95% confidence level. That, however, doesn’t mean that one can state that there is a 95% chance that more than 50% of the warming since 1950 was anthropogenic, as that falls foul of the prosecutor’s fallacy.

      Liked by 1 person

    17. Editor of the Fabius Maximus website Post author

      ATTP,

      (1) Your interpretation might be valid, but it’s not in the study. The study quite properly cautions about varied interpretations of much simpler language. It’s nice that you — like M Tobis in the thread about the pause — “know” how people answered surveys.

      (2) Note that neither Bart V (at his website) nor Tom Curtis (at SkS) raised your objection. Perhaps it was not so obvious or important to them?

      ** Thank you for the comments.

      Like

    18. ...and Then There's Physics

      It’s nice that you — like M Tobis in the thread about the pause — “know” how people answered surveys.

      That’s not what I’m saying, and I’ve no idea why you think that’s what I’ve said. For someone who was complaining about me not representing them properly, you seem to have no trouble doing that yourself. I’m pointing out that your suggestion that your analysis of the survey to show that only 47% accept that it is extremely likely that most of the warming since 1950 is NOT a correct representation of the consensus position. The consensus position is essentially that most of the warming since 1950 was anthropogenic. The “extremely likely” refers to the confidence in which we reject the null (that less than 50% of the warming since 1950 was anthropogenic). That it is extremely likely doesn’t change that the consensus position is still simply that most of the warming since 1950 has been anthropogenic.

      Note that neither Bart V (at his website) nor Tom Curtis (at SkS) raised your objection. Perhaps it was not so obvious or important to them?

      Tom Curtis’s point and mine are related. I’m not sure why Bart not raising the same point as me is some argument against my point. It would seem to suggest that there are many objections to what you’ve concluded with your analysis.

      Liked by 1 person

    19. Editor of the Fabius Maximus website Post author

      ATTP,

      “The consensus position is essentially that most of the warming since 1950 was anthropogenic. … That it is extremely likely doesn’t change that the consensus position is still simply that most of the warming since 1950 has been anthropogenic.”

      You in effect wash out of the statement any mention of confidence levels, so that the consensus is expressed as a “yes” or “no” agreement with the attribution statement. In which case the IPCC could have dispensed with the large apparatus of confidence levels, and the PBL survey could have been much shorter. However, in fact confidence levels are important inputs to public policy and washing them out of a statement of the consensus is not helpful.

      My conclusions

      (1) My guess is the PBL survey ended any interest in more surveys of this kind, due to the for desperate efforts we see by Bart, Tom, and others to back away from their broad survey of climate scientists and redefine “climate scientists” more narrowly to obtain a group that is suitably certain about their policy campaign. That might be the primary result of this research, and is in keeping with behavior seen in some previous research. We can check back in, say, 2 years to see if I’m correct.

      (2) I think we’ve reached the end of this discussion, and accurately determined each person’s views. Time will probably provide more insight about these things.

      (3) These points are literally academic, in the sense of theoretical interest only. The public policy debate is broken in the US. There will be no substantial US public policy measures taken in the foreseeable future, and hence probably none on a global scale (as the emerging nations will not act unless the US does). Weather will determine the evolution of public policy. All that remains is to see what lessons we can learn from this debacle. Details here.

      Thank you for posting your views in such detail, and taking the time to thrash out these things.

      Like

    20. ...and Then There's Physics

      You in effect wash out of the statement any mention of confidence levels, so that the consensus is expressed as a “yes” or “no” agreement with the attribution statement.

      I’m going to give up after this too, but think about this. The attribution statement is based on a standard hypothesis testing. The hypothesis is that more than 50% of the warming since 1950 was anthropogenic. This is tested by considering the null hypothesis (less than 50% of the warming since 1950 was anthropogenic). The null hypothesis is rejected at the 95% level, allowing us to use the term “extremely likely”. However, this does not change the hypothesis. The hypothesis remains as “more than 50% of the warming since 1950 was anthropogenic”. The “extremely likely” simply refers to the confidence we have in that hypothesis. We weren’t testing whether or not it was “extremely likely” – that is a consequence of hypothesis test, not part of what was being tested.

      The public policy debate is broken in the US.

      Yes, that does indeed appear to be the case and is not something that I applaud. It is of great concern that this debate is broken, both in the US and elsewhere.

      Liked by 1 person

    21. Editor of the Fabius Maximus website Post author

      ATTP,

      “The ‘extremely likely’ simply refers to the confidence we have in that hypothesis.”

      Yes, I am quite clear about that. And I agree, we’ve worked this thread as far as possible with the available information.

      “It is of great concern that this debate is broken, both in the US and elsewhere.”

      I agree. Hence the importance, IMO, of attempting to determine what went wrong. This is a vital but inherently speculative activity (i.e., we can ponder counterfacturals — paths not take). Your perspective as a scientists, but not a climate scientist, probably gives you a good perspective. You might find my guesses of interest.

      Like

  23. Pingback: PBL survey shows strong scientific consensus that global warming is largely driven by greenhouse gases | My view on climate change

  24. ...and Then There's Physics

    Chapter 10 of the WG1 discusses the attribution issue. On page 878, it says

    Attribution results are typically expressed in terms of conventional ‘frequentist’ confidence intervals or results of hypothesis tests: when it is reported that the response to anthropogenic GHG increase is very likely greater than half the total observed warming, it means that the null hypothesis that the GHG-induced warming is less than half the total can be rejected with the data available at the 10% significance level.

    Liked by 1 person

    Reply
  25. Pingback: More on Verheggen et al: Great Survey. Pity about the report… It’s still 66%. | The Lukewarmer's Way

  26. Pingback: PBL survey shows strong scientific consensus that global warming is largely driven by greenhouse gases | Enjeux énergies et environnement

  27. Pingback: The Global Warming Policy Forum (GWPF)

  28. Pingback: Weekly Climate and Energy News Roundup #191 | Watts Up With That?

  29. Pingback: How we broke the climate change debates. Lessons learned for the future. | Watts Up With That?

  30. Pingback: The conceits of consensus | Climate Etc.

  31. Pingback: The conceits of consensus | Enjeux énergies et environnement

  32. Pingback: How We Broke the Climate Change Debates. Lessons Learned for the Future | US Issues

  33. Shodo

    Ok… not a scientist. But your big headline catcher is NO CONSENSUS: Only 47% of scientists think that Anthropogenic GHG is the dominant driver of climate change”… Yet, when I read the Abstract of the study you are getting this from, this is what I read:

    “Results are presented from a survey held among 1868 scientists studying various aspects of climate change, including physical climate, climate impacts, and mitigation. The survey was unique in its size, broadness and level of detail. Consistent with other research, we found that, as the level of expertise in climate science grew, so too did the level of agreement on anthropogenic causation. 90% of respondents with more than 10 climate-related peer-reviewed publications (about half of all respondents), explicitly agreed with anthropogenic greenhouse gases (GHGs) being the dominant driver of recent global warming.”

    And all the graphs you post don’t back up your headline AT ALL…
    You are just simply playing around with the tables to twist it to say what you want it to say – For instance: in table 1b, in the “…more than 50%” side, if I add up the results from “More likely than not” to “Virtually certain”, the number of scientists who think that Anthropogenic GHG contribution is more that 50% is:

    33.6+31.6+24.1+8.3+1.8 = 99.4%
    … but you don’t want that to be your headline, do you?

    Like

    Reply
    1. Editor of the Fabius Maximus website Post author

      Shodo,

      The consensus tested here concerns the IPCC’s AR4 headline finding that “Most of the observed increase is global average temperature since the mid-20th century is very likely due to the observed increase in anthropogenic greenhouse gas concentrations.” Your math refers only to the first table; the AR4 finding is the result of both tables.

      Also note that later discussion about this revealed that the IPCC agreed with my finding. Buried on page 884 in Chapter 10 of AR5’s Working Group I:

      “We conclude, consistent with Hegerl et al. (2007b) {i.e., chapter 9 of AR4}, that more than half of the observed increase in GMST {global mean surface temperature} from 1951 to 2010 is very likely due to the observed anthropogenic increase in GHG {greenhouse gas} concentrations.”

      That lowered the IPCC’s confidence to 90%, matching the results of the PBl survey! As I said, this is well-done research.

      That’s below the 95% level standard for considering it significant. That’s important to know when evaluating Obama’s Clean Power Plan (details here). I support it, but the reasons concerning climate change are weak at best.

      Like

    2. Editor of the Fabius Maximus website Post author

      Shodo,

      Update: That was poorly written. I meant that they changed the headline finding from “very likley … anthropogenic greenhouse gases” to “extremely likely all anthropogenic forcings”. Few noticed this, hence all the headlines about the IPCC raising its confidence levels.

      The finding about greenhouse gases was moved to page 884. The important thing is that the PBL survey accurately describes the consensus about greenhouse gases as a majority of climate scientists at the 90% level (very likely), not the 95% level (“extremely likely”).

      Kudos to attorney Tom Curtis for discovering this. I don’t believe many scientists noticed this.

      Like

    1. Editor of the Fabius Maximus website Post author

      Aldog,

      Yes, there is a consensus that GHG causes warming. But that’s not the be all and end of of climate science. The IPCC is useful because it provides more specific — more operationally useful — finding. This post discusses the IPCC’s finding about the effect of anthropogenic GHG’s: how much warming, over what period, with what certainty.

      How can you have read the post and not understood specifically what consensus is being examined?

      Like

Leave a comment & share your thoughts...

Please log in using one of these methods to post your comment:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s