Arguments
The 97% consensus on global warming
What the science says...
| Select a level... |
Basic
|
Intermediate
|
Advanced
| ||||
|
97% of climate experts agree humans are causing global warming. |
|||||||
Climate Myth...
There is no consensus
The Petition Project features over 31,000 scientists signing the petition stating "There is no convincing scientific evidence that human release of carbon dioxide will, in the forseeable future, cause catastrophic heating of the Earth's atmosphere ...". (Petition Project)
Science achieves a consensus when scientists stop arguing. When a question is first asked – like ‘what would happen if we put a load more CO2 in the atmosphere?’ – there may be many hypotheses about cause and effect. Over a period of time, each idea is tested and retested – the processes of the scientific method – because all scientists know that reputation and kudos go to those who find the right answer (and everyone else becomes an irrelevant footnote in the history of science). Nearly all hypotheses will fall by the wayside during this testing period, because only one is going to answer the question properly, without leaving all kinds of odd dangling bits that don’t quite add up. Bad theories are usually rather untidy.
But the testing period must come to an end. Gradually, the focus of investigation narrows down to those avenues that continue to make sense, that still add up, and quite often a good theory will reveal additional answers, or make powerful predictions, that add substance to the theory.
So a consensus in science is different from a political one. There is no vote. Scientists just give up arguing because the sheer weight of consistent evidence is too compelling, the tide too strong to swim against any longer. Scientists change their minds on the basis of the evidence, and a consensus emerges over time. Not only do scientists stop arguing, they also start relying on each other's work. All science depends on that which precedes it, and when one scientist builds on the work of another, he acknowledges the work of others through citations. The work that forms the foundation of climate change science is cited with great frequency by many other scientists, demonstrating that the theory is widely accepted - and relied upon.
In the scientific field of climate studies – which is informed by many different disciplines – the consensus is demonstrated by the number of scientists who have stopped arguing about what is causing climate change – and that’s nearly all of them.
Authors of seven climate consensus studies — including Naomi Oreskes, Peter Doran, William Anderegg, Bart Verheggen, Ed Maibach, J. Stuart Carlton, and John Cook — co-authored a paper that should settle this question once and for all. The two key conclusions from the paper are:
1) Depending on exactly how you measure the expert consensus, it’s somewhere between 90% and 100% that agree humans are responsible for climate change, with most of our studies finding 97% consensus among publishing climate scientists.
2) The greater the climate expertise among those surveyed, the higher the consensus on human-caused global warming.
Expert consensus results on the question of human-caused global warming among the previous studies published by the co-authors of Cook et al. (2016). Illustration: John Cook. Available on the SkS Graphics page
Scientific consensus on human-caused global warming as compared to the expertise of the surveyed sample. There’s a strong correlation between consensus and climate science expertise. Illustration: John Cook. Available on the SkS Graphics page
Expert consensus is a powerful thing. People know we don’t have the time or capacity to learn about everything, and so we frequently defer to the conclusions of experts. It’s why we visit doctors when we’re ill. The same is true of climate change: most people defer to the expert consensus of climate scientists. Crucially, as we note in our paper:
Public perception of the scientific consensus has been found to be a gateway belief, affecting other climate beliefs and attitudes including policy support.
That’s why those who oppose taking action to curb climate change have engaged in a misinformation campaign to deny the existence of the expert consensus. They’ve been largely successful, as the public badly underestimate the expert consensus, in what we call the “consensus gap.” Only 16% of Americans realize that the consensus is above 90%.
Last updated on 8 May 2016 by BaerbelW. View Archives
LeonD - I expect you're doing a 'drive-by', rather than actually engaging in conversation, but I would ask you to consider just what proportion of peer-reviewed biology papers make explicit statements for or against the validity of evolution in their abstracts? And whether you, for some reason, think the large percentage of such papers not restating known facts is in some fashion disagreement with evolution?
The same holds of climate science. In fact, I suspect the estimated percentage of disagreement on climate is biased towards the negative (that the percentage might be lower than 3%), since authors disagreeing with the consensus have far more reason to mention AGW than authors who treat it as a known and understood background to the data.
Bzzzt.
The source is the paper itself:
http://iopscience.iop.org/article/10.1088/1748-9326/8/2/024024
Specifically Table 5
I am referring to the self-rated results but the abstract results are even less in favour of AGW.
LeonD - I'll repeat my question: do you think the high percentage of biology papers that fail to state a position on evolution are in fact evidence that biologists disagree with it? Or that the infinitesimal number of modern physics studies stating a position on the existence of atoms represents evidence of major disagreement there?
There's no need to repeat known facts, especially in the limited space of a paper or even more so the 200-500 words of an abstract - your argument is absurd.
I cannot find the reference at the moment, but as I recall Naomi Oreskes noted that as a scientific consensus grows the explicit mention of that consensus declines - because, again, there's no need to repeatedly tell your audience that water is wet, or that a clear sky is blue...
My mistake, I thought they were querying the authors on their own views not on what their papers were saying.
LeonD... I think that's a very common mistake. Relative to the Cook13 paper, many people fail to discern the difference between "position" and "opinion."
For a survey of scientific opinions, rather than the published work, see Doran 2009, whose survey found that among scientists who had more than half of their recent work on climate (i.e., who are actively researching the matter), 97% agreed that:
KR @707, I think insufficient attention has been paid to uncertainty intervals with regard to the concensus. In particular, in the case of Doran, Kendall and Zimmerman (2009), the sample size for question two, ie, the question on attribution, is only 77.
Calculating uncertainty depends not only on the sample size, but also (weakly) on the size of the total population. In the case of climate scientists, the total number of climate scientists in the world is an unknown. However, based on a literature review, Verheggen et al (2014) found the emails of approximately 8000 people, of which approximately 7600 where climate scientists (the other 400 being contacted because they where known "skeptics". On that basis, the total number of climate scientists in the world is likely to be greater than 5000, but less than 50000.
Using these figures and a confidence interval calculator, it is possible to determine that the 99% confidence interval is approximately is between + 2.6% and - 4.64 to 4.68%. The larger of the two figures assumes 50 thousand climate scientists. Of course the confidence interval calculator assumes a normal distribution, which is not possible in this case because there cannot be more than 100% concensus. That is likely to mean the lower bound is understated by a small amount, but the 95% confidence interval almost certainly has a lower bound less than or equal to 4.7% based on these figures.
More troubling for Doran is the actual question, which is:
By asking if human activity is "a significant factor", it allows that other influences are as, or even more significant.
Taking significance to be "statistically significant", it asks whether global temperature increase since the "pre-1800s" would have been less than that observed by a statistically significant amount absent human influence. Given the statistical uncertainty in determining pre-1800s temperatures (see graph below) that requires greater than 50% of the warming be attributed to anthropogenic factors. I think this means the question must be understood colloquially, where "significant" does not imply "statistically significant".
Colloquially, something contributing 25% of the effect would be considered "a significant contributing factor". Arguably something contributing just 10% of the effect would also be considered "a significant contributing factor" but that is more dubious. Taking the 25% benchmark, we can compare Doran et al to Verheggen et al, in which just over 90% agree that 25% or more of the warming is due to anthropogenic factors. Allowing for the inclusion of approximately 5% known "skeptics" without regard of their scientific qualifications (and in most cases absent relevant scientific expertise), that result is qualitatively equivalent to Doran et al's.
The upshot is that unless we are making the weak claim that the consensus is that anthropogenic factors are a significant factor in recent warming, we should no longer be citing Doran et al, and hence the 97% figure, for the percentage of scientists who accept the concensus position. That is particularly the case given Bray and von Storch (2010) and Verheggen et al (2014), both of which post date Doran et al, have larger sample sizes and support a consensus figure in the high 80 percents. In particular, Verheggen et al, excluding those invited because of their known "skeptical" opinion and without regard to their scientific qualification, find a concensus figure of 87% (85-89%).
Tom Curtis - I would agree that little attention has been paid to the uncertainty ranges on consensus estimates. However, as you yourself have noted WRT Doran, with perhaps the smallest sample, the uncertainty is <5% - meaning that even at the extrema we are still looking at a >90% consensus on AGW in the literature, and in at least some surveys of the expert opinions. (As I understand it, B. Verheggen is of the opinion that the lower number in their survey was actually due to a much more detailed/specific question, rather than the mean range thought appropriate - that the respondents didn't think they could narrow it down to the specificity given)
And when you look at actual attribution studies in AR5, the fraction of warming due to AGW has a mean of 110%, with less than a 5% chance of anthropogenic causes being responsible for less than 50% of observed warming. That makes AGW not just a significant, but a dominant cause.
Quite frankly, the various arguments on consensus (and denial thereof by the pseudoskeptics) are equivalent to discussing the number of angels who can dance on a pin, given that by any measure the scientific consensus on AGW is as high as that on ozone depletion by CFCs, acid rain, or the dangers of smoking tobacco, in all of which we found the consensus sufficient to act.
We know enough to take appropriate action.
KR @709, Verheggen et al argue that the percentage of respondents excluding undetermined results (ie, "unknown", "I do not know" and "other") for both the qualitative and quantitative responses are equivalent. Specifically, 84 +/- 2% of respondents agreed that 50% or more of "global warming since the mid 20th century" can be attributed to "human induced increases in atmospheric greenhouse gas (GHG) concentrations"; while 86 +/- 2% agreed that greenhouse gases had a moderate or strong warming contribution to the "reported global warming of ~0.8 degrees C since pre-industrial times".
As an aside, the unequal time periods for the quantitative and qualitative questions substantially weaken that argument. However, I think it is a no brainer that "I do not know" and "other" responses should not be included. On the other hand, arguably "unknown" responses claim scientific ignorance (ie, it has not been determined adequately by scientists) rather than mere personal ignorance, and so should not be included. Against that, an "unknown" response may merely indicate the respondent thinks it is not yet determined whether the greenhouse gas contribution was 75-100 or 100-125% (quantitative question) or a moderate or strong warming contribution (qualitative question). Therefore while presumable some respondents answering "unknown" do not agree with the consensus, it is problematic including the "unknown" figures because doing so assumes that all who so answered disagreed with the consensus which is not at all certain.
More important are the figures with no "unconvinced", ie, those deliberately invited to participate because of their "skeptical opinion" rather than because they are just scientists. Excluding both "undetermined" responses and "unconvinced" invitees, 87 +/-2% agreed that 50% plus of recent warming has been due to anthropogenic greenhouse gases. That does not lie in the uncertainty range of Doran et al. As Verheggen et al. is much more recent then Doran et al, we must therefore either conclude that there has been an approximately 10% slide in agreement with the concensus among climate scientists; or that differences in the questions made a substantial (approximately 10%) difference in the response. The later is what I argue above, based on the difference between "a significant contributor" and "the major contributor".
I completely agree with your final two paragraphs, but do not think that reason to by imprecise or selective when quoting determination of the size of the concensus. That is, to the best of our current knowledge, ~87% of climate scientists (on attribution), and ~97% of climate science papers. IMO those figures show that the approximately 13% of climate scientists who do not agree with the IPCC on attribution do not do so based on publishable evidence. Put another way, it means that political opinion has influenced the scientific views of some climate scientists, but against the IPCC position, not for it (ie, in the opposite direction of the bias claimed by "skeptics").
Response to sjw40364 on the appropriate thread.
Okay so while a sceptic is mostly interested in checking (and if necessary refuting) new scientific claims, it is reasonable to discuss the "consensus" issue due to its importance to science as a whole.
Ironically, there is no consensus on the meanings of the terms used to define this consensus. Does it mean a majority? Or just an important and strongly agreed minority? Do voices with authority carry sway, or is it democratic. Does it need to be "overwhelming" and is it, in fact? Is it absolute or is there internal dissent? Which human beings count as scientists? Which institutions act as gatekeepers thereof and what is their motivation?
Rather than work trough all of these, I will simply ask the reader to consider whether it is healthy that you are being asked to accept the speakers' tacit definitions on these matters, as well as their unstated assumptions. You are being guided toward what is really more of a psychological sensation than any fact-based argument - the sensation of being part of something big and powerful. Maybe a sense of belonging and safety. Maybe moral superiority. Maybe the clarity that comes from being decisively led.
If the reader is ready to understand their own (and their peers') fralties in such areas, then I do not need to discuss the history of systems of control and subjudication. If not, there's no point getting in to it except to suggest you may wish to begin with the Milgram experiment.
Instead, I will take a single example, from the current article, of a flagrent manipulation of the meaning of "consensus" and surrounding terms: the 97% pie-chart.
You thought it said 97% of scientists, right? No. It's 97% of papers. That's right there in the jpeg image itself but you didn't notice it. What else didn't you notice?
If you read the underlying study, what 97% really agree with is somthing along the lines of "do you agree that (a) humans emit CO2 and (b) that the greenhouse effect is real". Your present author does, and so would be a part of the consensus!
The trick here is a toxic mix of pedantry and tactical naivety - as so often seen among precotious fifteen-year-olds, but in this case carefully hidden within a typically dull metholodgy section in a paper. It is *pedantically* true that human CO2 plus greenhouse effect implies *probably* *some* human generated warming. But has human generated warming been *shown* to occur? Not implied. Is it problematic? Not implied. Significant? Not implied. Even detectable? Not implied. Nor does the position in the question even imply that there won't be compensating factors or that warming would even be harmful anyway.
In summary, consensus taken in general is too subject to the frailties of the human condition for any wise person to pay any attention to it. Specific factoids, such as the 97% pie-chart (and there are others) may seem to lend concrete validity, but as soon as you check them you find nothing meaningful, only trickery.
Should we accept climate consensus because consensus exists around, say evolution? A real sceptic can answer this easily: the whole point of science is to investigate methodically the questions whose answers are *not* obvious on the surface. No scientist would ever be so intellectually lazy as to reason that since the climate consensus "sort-of looks like" the Darwinian consensus, that their underlying scientific validities should also match.
Climate consensus is much more like a rainbow. Amazing to look at; vast and magical. But how many times do you have to check for that pot of gold before you accept there's really *nothing there at all*!
A.R.S.Says @ #712 :
To be blunt: the word consensus has a very plain, straightforward meaning in the English language.
Your expressed "logic" is a complete failure, since you seem unable to connect words and concepts and realities.
(btw, I must commend your sense of humour in choice of your nom-de-plume ~ the abbreviation is priceless.)
[JH] Inflamatory & off-topic.
I'm looking for the data on climate change. I haven't been able to find it. I don't care how many agree, I want to know what evidence they base their opinion on. I was taught to question, not swallow. Please just post the data on climate, not the politics of popularity. Thank you.
[DB] Data and codes are openly available, and have been for years:
http://www.realclimate.org/index.php/data-sources/#Climate_data_raw
http://www.realclimate.org/index.php/data-sources/#Climate_data_processed
http://www.realclimate.org/index.php/data-sources/#GCM_code
http://oco.jpl.nasa.gov/science/dataproducts/
http://podaac.jpl.nasa.gov/
http://daac.gsfc.nasa.gov/
https://earthdata.nasa.gov/about/daacs
https://www.wmo.int/pages/themes/climate/climate_models.php
Note that the Muir Russell Commission was able to do a full global reconstruction from the raw data linked to from the above page, WITHOUT ANY CODE, in a mere 2 days (when asked, they replied "any competent researcher could have done the same).
The Auditors over at McIntyre's Climate Audit have been struggling with their "audit" reconstruction for many years now.
KiAnCa @ #714 : You have my sympathies, for your desire to gain good quality scientific information about the amount of global warming going, and how severely the problem is building. As you have doubtless already noticed, the mainstream media generally does a poor job in supplying realistic information ~ and it gives an inordinate amount of space to anti-science propagandists (with lawyer-type rhetoric designed to make you think black is white, or that there are so many "doubtful" shades inbetween, such that there is nothing meaningful in this whole universe).
You will find a vast amount of science-based info on this website . . . but you have my sympathy, because that info is not presented as a giant-size single meal where you simply chew your way through from one end of the pie to the other end.
Best (a) if you go to the Home page, central top region, and click on the small box titled "The Big Picture" . . . and then follow to areas that interest you,
or (b) on the Home page, click on the nearby box titled "Newcomers, start here" . . . and look down to the second heading, titled "Good starting point for newbies" where [second line of the paragraph] you can click on "Warming Indicators" and from that go to "Evidence for Global Warming (intermediate)" . . . where you can follow your interests. I must point out that this particular section generally holds info up to about 2010 ~ and so doesn't directly mention all the additional weight of scientific info in the last five years [i.e. all the newer "hot year" global records and even faster Ice Melt and sea-level rise]. No great matter, since the "sufficient evidence" was already overwhelmingly convincing, long before that date [indeed, in a recently publicised scandal, it appears that Exxon already had convincing evidence of the CO2/Warming problem by 1979],
or (c) if your scientific education is already above average, then you can simply skip to the "Arguments" [on Home page] and pursue any of the 170+ "arguments" [arranged by Climate Myth] which interest you, and delve further from there. The Myths are quite entertaining, because the info there does neatly deflate all the rubbish/nonsense talked by the small number of shills & mavericks who oppose the mainstream science (i.e. the mainstream science which results in virtually all the climate scientists being in accord with the consensus of 97% . . . or nowadays more likely 99% )
Good hunting ~ and please use the appropriate thread's Comments Section for any questions that you want clarified.
One of the human finger prints cited in the first week of the Denial course was that the atmospheric warming this century is unique in the fact of warming lower atmosphere and cooling upper atmosphere. What evidence from past warming episodes establises that this is unique to the current warming. How do we know what happened in the upper stmosphere in the past warming/ increased CO2 events?
[Rob P] - A cooling upper atmosphere and warming lower atmosphere is a signature unique to the enhanced (increased) Greenhouse Effect. If we had a Tardis, we would be able to go back in time to the Paleoecene-Eocene Thermal Maximum (PETM) about 55-56 million years ago, a time of substantial natural global warming, and observe the Greenhouse Effect growing stronger.
The enhanced Greenhouse Effect we are now measuring is a human fingerprint because the source of it is the continued emission of greenhouse gases, primarily carbon dioxide, produced by industrial activity. See the SkS post: Climate Change Cluedo.
Thanks for Responce @Rob P and for the link to Climate Cluedo. I get that carbon isotopes are critical in determining CO2 sources and ways of determining concentrations but my question speicifically was what is it about a cooling upper atmosphere in conjunction with a warming lower atmosphere that is unique. Another way to ask this might be, why is the upper atmosphere cooling with increased GHG levels while the lower atmosphere continues to rise at a sharp rate compared to background seasonal oscilations? And how do we know that in the past when the lower atmosphere warmed, so to did the upper atmospthere, or did it just stay the same. (I only found three hits on the Cluedo page when searching "upper atmos" and they were all in comments. no hits for "lower atmos")
What you should be looking for is "stratospheric cooling". It is not an easily understood concept, but there are several attempts around the internet to explain it. At basic level, It falls out of the equations for radiative transfer if you increase a greenhouse gas. Other forcings that change the surface temperature like changing albedo, solar influx, or aerosols do not produce this effect.
John writes: "Science achieves a consensus when scientists stop arguing"
Actually that's a bit simplistic. A scientifi consensus is formed after a series of scientists are able to reproduce the work of the scientist advancing a hypothesis. This is done by publishing confiming/denying results in refreed journals. Tom make that possible, the person advancing the hypothesis first fully explains it, then describes how it was tested (the "mehtodology"), the observed data and the results.
A scientific consensus isn't formed by simple agreement between scientists, it's evidence based and very much dependent on repeatable experiment. So while the consensus that CO2 is a "greenhouse" gas, meaning that like water vapor and methane it absorbs and radiates solar energy in known quanta, there is no consensus on the effect or "sensitivity" Earth's climate has to increases or decreases in it. Which is the problem.
We know CO2 absorbs IR. Water vaport (H20) observes much more, so much more that IR astronomers put their telescopes as high as possible, on Mauna Kea, Medium Altitude soborbital platforms like the KAO and SOPHIA, and in low Earth orbit in order to get above H20. IR astronomers aren't particularly worried about CO2 because its effect is so small it just doesn't matter.
[PS] Myths about water vapour are addressed under "water is the most powerful greenhouse gas". Make your arguments there. Offtopic comments will be deleted.
"Tom" doesn't make it possible. "To make"
"methodology" not "mehtodology"
"vapor" not "vaport"
"suborbital" not "soborbital"
I was proof reading my post here on the last page of comments when I encountered this gem:
"One of the human finger prints cited in the first week of the Denial course was that the atmospheric warming this century is unique in the fact of warming lower atmosphere and cooling upper atmosphere"
Not sure who came up with this but it's trully choice. So how many folks were measuring the temperature of Earth's stratosphere 200 years ago? 500 years ago? 2000? 20,000 years ago?
Whoever made up that fun fact should get a prize, it's a real whopper.
[PS] try reading for understanding rather than demostrating misunderstanding before banding about accusations. The surface temperature of any planet can be altered by changing solar input, albedo, GHG composition or aerosols. Increases in GHG composition is unique in that it is only forcing change that will warm the surface but cool stratosphere.
It would be very nice if this site allowed comments to be edited.
@Pfc. Parts
I gather from you're tone your here to troll not to learn, understand or convey science. But if you're interested in the source you can listen the full interview with Ben Santer (lead author of the historic 1995 IPCC) here:
https://www.youtube.com/watch?v=EOrUYQhGzT8
He states, no known natural mechanisms or combination of natural causes have that sustained effect, that human fingerprint, in this unquie way.
to Pfc.Parts @722 : you make a fair point, with your comment "It would be very nice if this site allowed comments to be edited."
On balance though, that would not be a good idea ~ and I am sure you can picture the chaos and non-sequiturs which would occur as posters go back and modify their posts, even with innocent intent (let alone the malicious intent). Nope: to be fair to all, a non-self-modified posting system is definitely far better.
Mind you, it could be reasonable to allow a poster to later insert a very obvious "corrigendum" paragraph at the end, to deal with clumsy bloopers / typos / or poorly-expressed phrasing . . . and this would help the flow of understanding in the commentary [rather than having such corrections appear later and quite possibly be half-buried by other intervening posts]. Such corrigendum would require clear demarcation and date/time label.
But . . . there would probably need to be a 24 or 48-hour cut-off for such "grafted-on" corrections. And even there, I am sure you can picture how some posters would try to play games and thoroughly abuse such a system. So, overall, it's simpler to keep this as they are : and it also makes for a simpler and less vulnerable control of the comments column.
[apologies for this off-topic excursion]
If you read the sentence stating 97% support, it's a self selecting subset of the data,
"of papers stating a position on human caused global warming"
of all the papaers in the Cook study, only 0.5% Explicitly support and quantify AWG as > 50%., (64 out of 11944)
of all papers stating a position, that number jumps to a whopping 1.5%. (64 out of 3974)
can someone explain to me how that equates to "consensus"