Okay, Climate Scientists: Time to Fight Smart

By Chris Mooney | November 8, 2010 8:55 am

The big news today is that climate researchers are banding together, preparing for the onslaught they fear is coming at the hands of the new Congress. I’m honored to learn that one of the researchers behind the effort, Scott Mandia, was inspired by our book Unscientific America:

The science of climate change and even the scientists themselves are under attack from a well-orchestrated and well-oiled misinformation campaign.  The best defense against this anti-science offensive is to make sure that the correct message reaches a wide audience.  Chris Mooney & Sheril Kirshenbaum in their book Unscientific America: How Scientific Illiteracy Threatens our Future explain that scientists have failed to get their message across for a variety of reasons but mostly because we are not engaging the public on their turf.  After reading that book, I became a climate change evangelist with my Global Warming: Man or Myth? Website, this blog, and more recently a Facebook Fan Group called Global Warming Fact of the Day.  I have two small children and I do not like the future that I see for them or for their children in a human-driven warmer world.

Go Scott–but if we’re to be out there, we also have to be smart. The media is not at all like the scientific world. There are great potential benefits that can come from outreach and being outspoken–but they don’t come automatically. They’re the consequence of very deliberate choices about how to communicate effectively.

Some things to consider: When it comes to science communication, the facts are the baseline from which one absolutely cannot stray; but at the same time, we have to be aware that people respond most strongly to the frame. Climate scientists are going to be rebutting political attacks on them, but what will their message be? Will it be that incoming Republicans (and Attorney General Cuccinelli) are abusing their power? Or will it be that climate research is robust and convincing on the subject of anthropogenic causation?

Or, will it be that this is all a distraction from the real issue, which is to get clean energy solutions hopping before countries like China establish so big a lead that we’ll never catch up?

Notice that only one of these messages is purely scientific in nature. I suspect that’s the message that scientists will want to stick with–appropriately enough–but even here, there are pitfalls. Remember that the political attack is also largely scientific in nature, at least in terms of its framing. It exaggerates uncertainty about particular scientific studies (like the Hockey Stick) in order to distract from the big picture.

So any scientist walking into this context had better be ready for one obvious trap: Being lured into talking about uncertainty to the detriment of what we actually know. It’s easy to ask a scientist a question that will invite a large volume of caveats and doubt-generating statements without leaving much time to discuss what’s firm, what we can rely on. A question like, “what are the limitations of existing climate models?” You get the picture.

Conversations about uncertainty invoke a frame–a scientific frame–and that’s often precisely where climate skeptics want you to go. That’s where they live. That’s where they’ve been living for years: Just read Naomi Oreskes’ and Eric Conway’s book Merchants of Doubt. And they can get you into their territory just by asking you to be frank about your research.

That’s one of many things to bear in mind as we begin communicating climate science with a much greater public focus, and in a highly politicized context.

Another concerns the subject of advocacy. Scientists are very scrupulous when it comes to respecting the line between stating the facts, and pushing for policy outcomes–as they should be. But guess what: When you’re out there communicating about science in a politicized context, you’re going to be accused of being an advocate no matter what you do. It’s a false charge, but it’s a virtual guarantee that it will be made anyway.

It simply has to be brushed off. Don’t get angry, and don’t get distracted. Remember what it is that the American public, and political leaders, need to know about climate research–and tell them that. Tell them twenty times. And then do it again.

As this all unfolds, I’m looking forward to many further exchanges with members of the climate science community on how to proceed. It’s a challenging time, but also an exciting one. We can do this much better than we’ve done before. I’m glad the climate science community is ready to take a much needed step to reach out to a public that needs it very badly indeed.


Comments (21)

  1. RK

    Prior to mankind ever existing, there are scientific examples of massive climate change and major extinction events. Skipping the “hit” of a large object near what would become Central America 65 million years ago, it is clear that massive climate changes occur over time whether there are humans on the planet or not. Yet the article above, while talking about how we have to do a better job communicating science, quotes and complements a non-scientific opinion statement as though it was fact: “I have two small children and I do not like the future that I see for them or for their children in a human-driven warmer world.”

    There is no mention in the article of how some raw data used for climate models has been reported as “missing”, or how some data collection sites may have skewed data from being too close to man made structures that affected temperature collection. And the fact that funding for studies seems to follow those that proclaim a man made climate emergency…funding that would not flow if it would appear otherwise.

    There are examples that bias has entered the scientific process, both in the collection of data, and the people who process that data.

    So sure, lets have an examination of the science of climate change…but lets strip out the skewing of data and realize there is some element of self interest on the part of those who proclaim “its all man’s fault”. More openness is the answer here, moreso than usually required due to the bias that has crept into the scientific process.

  2. Chris,

    This is all good advice, but it’s advice that you should also consider. By continuing to push the “Republican War on Science” metaphor, you’re fueling the ideological heat surrounding the issue and distracting from substantive discussion of politics, policy, and science. You’re also stoking the same raw emotion and anger that you warn scientists against. It’s self-serving, distracting, and harms efforts at broader public engagement.

    As passionate as many of your readers might get when thinking about the “looming Republican war on science,” as Sheril put it the other day at this blog, the broader public simply does not care and is likely to see it as just more elite bickering, especially in the context of a economy where now a majority of Americans say they are concerned about meeting next month’s rent or mortgage payment.

    Consider the findings of last year’s Pew/ AAAS survey. Despite all the attention within the elite press this past decade to politicization of science by the Bush administration, only 10% of the public reported hearing a “a lot” about the issue compared to 55% of AAAS members. And only 17% of Americans said they thought that cases of politicization in the Bush era were more prevalent than under previous Administrations.


  3. Bobito

    @Chris “you’re going to be accused of being an advocate no matter what you do. It’s a false charge, but it’s a virtual guarantee that it will be made anyway.”

    Specifically “It’s a false charge”. Just changing “It’s” to “may be” would make a huge difference in communicating with the public. You are displaying bias by implying that any charge of “advocacy”, against a scientist, is a false charge.

    And this “Scientists are very scrupulous when it comes to respecting the line between stating the facts”. Again, it’s as if you are saying scientist are not capable of being dishonest.

    Is it that difficult to conclude that, whether consciously or unconsciously, climate scientists could have a bias toward their work? And may have a bias towards a political agenda as it will gain them more funding to continue their work?

    Scientist are people too, fallible people…

  4. Michael

    Sigh…the AGW folks never say never, even when the evidence is overwhelmingly against them. Well, so be it.

    People now know that the Man-made Global Warming fiasco is the grandest scientific hoax ever propagated. The Earth continues to get wetter and cooler, and livestock keep dying and crop yields keep getting smaller and smaller. Food prices continue to climb because of the global cooling.

    The real tragedy here is the extreme damage all of this global warming nonsense is doing to the credibility of science in general. Science is so absolutely vital to the well-being of humans and other species, and it’s a horrible shame to see it’s reputation slammed by the blind fanatics of man-made global warming.

    However, a stopped clock is correct two times a day, and after the cooling period there will be yet another warming period as the cycle continues…

  5. @Michael I’m curious about the evidence for climate cooling. Do you have peer-reviewed science literature on the topic?

  6. RK

    Roger Harris…take a look at http://www.telegraph.co.uk/earth/environment/globalwarming/6636563/University-of-East-Anglia-emails-the-most-contentious-quotes.html

    This is an example of why people are looking at global warming/climate change “scientists” with a strong degree of suspicion.

  7. @RK Thank you, but am not asking for evidence of a debate, of which I am well aware. I was asking for actual scientific papers in peer-reviewed journals that present evidence of a cooling trend, as asserted by @Michael. That way, perhaps we can assess the data for ourselves rather than relying on a semi-anonymous blog commenter.

  8. Nullius in Verba


    The concept of a ‘trend’ depends on having a particular statistical model of the data’s behaviour. Pick one model, and you will get one trend, pick another model, and you get a completely different trend. In this case, people are fitting a particular statistical model to data which it doesn’t fit, and so the result you get depends on what time interval you choose to apply it to.

    So for example over the past couple of months the trend is down, over the past year, it is up, over the past five to ten years it is down, over the past thirty years it is up, over the past hundred years it goes up and then down and then up, over the past thousand years it probably goes down but with a big dip in the middle, over the past ten thousand years it goes distinctly down. Over the past twenty to a hundred thousand years it goes strongly up, and over the past hundred million years it goes even more strongly down.

    So what’s “the” trend? A meaningless question, I suggest. But my guess is that Michael was referring to Professors Easterbrook’s and Latif’s prediction in this story: http://news.bbc.co.uk/1/hi/8299079.stm

  9. david arnold

    one of the best awakening graphic “pictures” i have seen of late is a chart that compares actual measurements of declining arctic sea ice with what the ?dozen? major computer models predicted. no contest. the melt significantly is outpacing the most dire forecasts. (alas, i cannot pull up the url for reasons i will not bore you).

    “just the facts, m’am”

  10. Nullius in Verba


    Yes, that’s because the melt is predominantly caused by a change in the prevailing wind direction, pushing ice southwards into warmer waters. The computer models only modelled temperature. They didn’t predict the change in wind patterns and so underestimated the melt.

  11. Marco

    9+11, the Daily Mail article was a good example of the philosophy of “you will be accused advocacy, no matter what you do”. Jones states that there was no statistically significant since 1995, which merely means that the error bars are larger than the trend. Now, 1995 was very nicely selected, since it WOULD have been significant if the starting year was 1994. More gríevous, and in some way a communication error by Jones, is that the analysis shows there is a 5% chance that there was no warming since 1995, but an equally large chance that the warming was TWICE AS MUCH as the actual trend.

    The BBC story NiV cites unfortunately manages to completely garble Latif’s “predictions”, thanks to Fred Pearce doing a bad job on reporting what Latif said (the BBC most certainly did not contact Latif). Latif was reporting on a paper he was co-author on, Keenlyside et al, which “predicted” a 10-year period of little warming. Not cooling, little warming. The propagation of error in a nutshell, and another example that scientists, when using their accurate “science speak”, still get their words twisted to mean something else.

    Don Easterbrook is the laughing stock of actual scientists:
    Sadly, it has to be said. It’s the “emeritus” syndrome, apparently.

  12. Nullius in Verba


    I don’t think the Daily Mail article accused Jones of advocacy. Did you mean somebody else?

    On the whole, sceptics have relatively few complaints about that BBC interview – it was the first time that Jones responded non-confrontationally and openly about disputed issues, and it did his reputation a lot of good. It was the first (and to date about the only) thing they did right after Climategate.

    It is quite correct that 1995 was selected precisely because this was the limit of statistical significance. That was the point. The aim was not to try to use this to claim it disproved AGW or that temperatures had not risen in the interval 1975-2000 – most sceptics have a more sophisticated understanding than that. It actually arises from a question of whether global warming is falsifiable. A hypothesis that cannot be falsified, because any conceivable contradictory observations can be explained away as a temporary blip, is not scientific. You have to set a threshold beforehand at which you would say “hang on, the hypothesis of continued warming must be wrong.” So the question arises, how long does a negative trend, or lack of trend have to last before the hypothesis is falsified? How long do ‘blips’ typically last? And then we ask how long has the observed lack of significant trend actually been? So long as the latter remains shorter than the former, GW has not been falsified, but the longer a period you demand, the greater the risk that the preceding 25 years of increase will be judged on the same grounds to be a ‘temporary blip’ too.

    The significance of Jones admission was that up until then climate scientists had been refusing to acknowledge that there was any observed short term lack of significant trend – apparently in an attempt to avoid what they and you assumed the sceptics were trying to do. The attempt to deny a simple and straightforward result to avoid handing sceptics ammunition handed them a huge piece of ammunition because their denial of the obvious gave sceptics an easy way to refute what they said. By acknowledging the point, Phil Jones took that whole line of attack away.

    “More gríevous, and in some way a communication error by Jones, is that the analysis shows there is a 5% chance that there was no warming since 1995, but an equally large chance that the warming was TWICE AS MUCH as the actual trend.”

    No, the test Phil used doesn’t work like that. It takes a null hypothesis of no trend, and asks how likely an apparent trend bigger than that in either direction is to happen by chance, and concludes that there’s a 2.5% chance of a positive trend greater than that, and a 2.5% chance of a negative trend less than the negative of that. It says nothing at all about the probability of a trend twice as big. (Which I assume is what you meant. The way you phrased it was a bit garbled: – there’s absolutely no doubt that there has been warming observed since 1995, the question is whether this is distinguishable from random weather.)

    In fact, Phil’s test uses a noise model called AR(1) for it’s zero-trend null hypothesis; if you picked a different noise model like ARMA(1,1), which fits the data much better, then the confidence interval is even wider, and you would have to go back some years before even 1994 before it was significant again. I’m not complaining about the way Phil Jones calculated or presented his result – just noting that it depends on the assumptions you make.

    I would like to note that Deltoid is the laughing stock of actual scientists, and you shouldn’t assume that Tim’s opinion carries any weight with anyone on the other side of the debate. You’re welcome to put your own trust whatever authority figures you like, but you should be aware that it’s a waste of time using it as an argument to try to persuade anyone who doesn’t. There are plenty of pro-AGW scientists who are respected by sceptics. Quoting Deltoid as an authority on anything will just strike people as funny.

    Much of the problem here is a failure to communicate leading to misunderstanding. People on either side of the debate often don’t mean quite what the other side thinks or assumes they mean. It would help if the two sides listened more.

  13. Marco

    NiV: the Jones interview and the abuse of Jones’ quote is an example that whatever you do, your words will get twisted. This is the same mechanism as being accused of advocacy, whatever you say.

    re. Jones: There is no solid scientific reason to take zero-trend as the null hypothesis, it merely was the implicit null hypothesis in the question. Jones should have recognised that. He could easily have reworded the statement and tested against the previous decade’s warming rate, and he would have found it was not significantly different from that either.
    Your claim that climate scientists did not admit before there was no significant trend is without any evidence provided. Not surprising, because there is no such evidence. Many have pointed out that it is easy to find periods of “no significant trend” in the temperature record, even in the period of the 1980s and 1990s, where the 10-15 year trend is around 0.3/decade.

    Finally, Tim Lambert quite aptly shows how Don Easterbrook, notably in a lecture in which he accused other climate scientists of fraud, *deliberately* left out information. The “0” anomaly line Easterbrook used he referred to as “present”, without telling his audience that that “present” was more than 60 years ago. I’m not going for the “authority” here, I am going for the facts. I wish more people would do so.

  14. Nullius in Verba


    Whether the Jones quote was being abused depends on what you think it was being used for. It also depends on whether you happen to agree or disagree. But it is in the nature of controversy that the two sides disagree on the proper conclusions to draw, so if you use this definition then it is inevitable that every statement made by either side on which there is disagreement over how it should be interpreted will be “abused”. It becomes an empty definition. It becomes just an emotive way to say you don’t agree with it.

    Regarding the null hypothesis, it was as you say implicit in the question. If you choose not to answer the question, then you can of course pick a different null hypothesis, but what would that achieve?

    When you say, “Jones should have recognised that”, Jones did recognise that. But the entire point of the interview was to show a different face to the one seen in the Climategate emails; to show that he could answer a straightforward scientific question without having to weasel around with it so as to advocate for a particular interpretation. It’s precisely that sort of thing that got him into trouble in the first place. In saying that Jones should have recognised the trap and answered a different question, you just repeat the same error of substituting partisan advocacy for science.

    To a large extent, it was the exposure of that sort of thinking that made Climategate such a big deal, but since I assume you believe the partisan advocacy is science, I suppose it isn’t surprising that you don’t see it. So long as you’re not thinking “what is the answer to the question?” but instead “how can I phrase this so that it can’t be used to support an interpretation I don’t agree with?” the situation (and the attacks) will persist.

    You say that I didn’t provide any evidence, but fortunately you provided some for me! Tim Lambert’s page that you linked makes precisely this argument. (Thanks!)

    The simple fact is that fitting a linear trend to data is only meaningful if it’s behaviour consists of a linear trend plus statistically independent measurement errors with a zero mean Gaussian distribution, and weather data does not behave that way. Thus, if you choose one interval you get one answer, if you choose a different interval you get a different answer, which surely makes it obvious that neither is the right answer. It’s just random junk being pumped out of an algorithm. It’s the GIGO principle.

    But to the advocates, it’s quite clear that only intervals that result in a positive trend are “right”, and that intervals that result in a negative or zero trend are “wrong”, and that only the “right” trend lines should ever be shown. (You have to pick cherries if you want to make cherry pie, as they say.) Anybody who shows one of the “wrong” trend lines, on the other hand, is being deliberately deceptive. Statistical manipulations that increase a positive trend are perfectly legitimate clarification (a useful “trick”), but if you use exactly the same methods where they emphasise a negative trend, that’s just dreadful.

    In this particular case, Easterbrook evidently fitted a trend line to the smoothed data, which I agree is a poor technique. But climate scientists have done exactly the same thing on other occasions (often not even bothering to show the raw, unsmoothed data) and seen nothing wrong with it.

    The problem is, Easterbrook is using their own techniques and showing that he can get opposite answers – but what he ought to do is to explain that this shows that they don’t mean anything, not to substitute “the world is cooling” for the equally false “the world is warming”. (Or make it clear that by “cooling” he is not referring to a linear trend.) But Lambert completely misses the point, of course, and continues to assert in his conclusion that “the world is warming”. That was exactly what I was referring to.

    I too wish more people would go for the facts. If you had said you knew of an example where Easterbrook had drawn a trend line to show cooling, but that had he included data from the future after his lecture has been given it would have shown warming, then I’d have probably agreed with you. I had just said that trend lines are junk, so if you had just shown me a trendline Easterbrook had produced, from smoothed data no less, I’d have been forced to agree. But by citing Deltoid, you gave me the opportunity to instead talk about all of Lambert’s errors.

    I take your point that you intended to refer to the content of the post for your demonstration rather than the authority of its author, but you did say “Don Easterbrook is the laughing stock of actual scientists” so I presumed “actual scientists” was a reference to Lambert. As a university professor, Easterbrook is an “actual scientist” too, as are many sceptics, and you can’t switch definitions part way through a sentence. Whether or not he is a “laughing stock” does depend on the author you cite – and if you want to make a persuasive argument, then I was just saying that Lambert is a poor choice for that.

  15. Marco

    NiV, give me one example where Jones shows partisan advocacy. Note that I can come with one example, one that points to a paper being hopelessly flawed (which it was, and after the correction still is), by two authors that are doing partisan advocacy. If there is any example of GIGO, their paper is exemplary!

    Regarding the trend: Jones should have noted that there was a very strong El Nino in the early part of the chosen time period, and a very strong La Nina in the last part of the chosen time period, and that this would likely generate a negative slope if there was no underlying warming trend. The framing of the question is deliberately jinxed, and the null hypothesis being a zero trend is actually an example of a partisan advocacy standpoint!

    Finally, Easterbrook is currently behaving in a way that is far from scientific, and hence has left the realm of the scientists. Your claims that others used statistical tricks to increase a positive trend are without any evidence. I’m not surprised…

  16. Nullius in Verba


    “NiV, give me one example where Jones shows partisan advocacy.”

    Hide the decline.

    “Jones should have noted […] if there was no underlying warming trend.”

    Jones should have simply answered the question. To a scientist, the evidence is the evidence, and you report it all – the bits that favour your hypothesis and the bits that go against it. Especially the bits that go against it, because those are the parts that really test the hypothesis, to confirm or falsify. To do so is the mark of a genuine scientist. Also, if we know (or suspect) that you’re not going to mention contrary results, it destroys the credibility of the results you do report. The fact that Jones spoke as he did on this occasion was impressive, and helped his credibility.

    The fact that you’re still sat there trying to come up with points and arguments to prove that it somehow doesn’t count shows that you’re still missing the point. Jones was not trying to “win” the argument against the sceptics, he was trying to rebuild his reputation as a scientist. As such, it was absolutely essential that on being asked a factual question that he give a straightforward factual answer. Hard as it may be for you to believe, doing so is actually more convincing than trying to dismiss or downplay possible alternative conclusions.

    Global warming, defined simply as an increase in the average temperature of the globe, can temporarily start or stop without contradicting the hypothesis that CO2 is contributing significantly to global temperatures. The term “global warming” does not refer to the “underlying trend” contributed by CO2; they are not in any way equivalent. Saying that “over the past 15 years global warming has stopped” does not disprove AGW, and we all know that. If any sceptics say so you can win that argument, and most sceptics have learned not to. But denying or avoiding mention of easily verifiable facts purely to avoid giving sceptics ammunition has precisely the opposite effect. First, we can show you unambiguously tinkering with the facts to try to make your case. And second, it makes everyone wonder what else you’re carefully not mentioning.

    You are correct that Jones could have mentioned the contribution of ENSO on the record. But if he was going to do so, then he ought also to mention the effects of the AMO and PDO on temperatures. If you’re going to move on from the fact of warming or cooling to discuss potential causes, then you have to discuss them all, and not only when it suits you. In this case, they’re not relevant to the question that was asked, so to selectively include them and not the others would have simply looked like falling back into advocacy again. Precisely the opposite of the effect intended.

    “the null hypothesis being a zero trend is actually an example of a partisan advocacy standpoint!”

    I really don’t know what to say to this. I’m not even sure what you mean by it, but it seems to me so far outside the frame of science that I can’t believe it’s what you really intended to say. The null hypothesis is what you are trying to disprove. The trend being zero is precisely what we keep being told you have “mountains” of evidence to be able to dismiss. Of course you’re going to use a zero-trend null hypothesis. If you didn’t, then the possibility of a zero trend would remain open, and the sceptics would win.

    I have no intention of defending Easterbrook, and have not done so. We were asked what Michael was referring to, and I told you. In the original case, Easterbrook’s argument actually sort of helps AGW, so I’m not clear on why you’re trying so hard to discredit him. I also said that linear trends up or down were meaningless here, so to the extent that we’ve diverted to this alternative case that Deltoid commented on, I am not disagreeing. But if you define scientists as people who behave scientifically, then those climate scientists are not scientists either – to a far greater extent than Easterbrook. Nobody’s perfect.

    Evidence for my claims that others have used statistical “tricks” to accentuate positive trends and “hide” negative ones is copious and easily available. It is simply not credible that you can have been long involved in this debate and not be aware of it. My comment above was already too long, I’ve explained it at length before on this blog, and I don’t have the space to recite every time what everybody here should know already.

    If you genuinely don’t know, then for a polite request and show of open-mindedness I can give you lots of pointers and reading recommendations. But I get the feeling that you simply don’t want to know, and will interpret anything you see so as to retain belief in your predetermined conclusion. If you’re confident that I cannot possibly be right, then no need to ask. I believe in freedom of belief – your beliefs, and your reasons for holding them, are your own business.


Discover's Newsletter

Sign up to get the latest science news delivered weekly right to your inbox!

About Chris Mooney

Chris is a science and political journalist and commentator and the author of three books, including the New York Times bestselling The Republican War on Science--dubbed "a landmark in contemporary political reporting" by Salon.com and a "well-researched, closely argued and amply referenced indictment of the right wing's assault on science and scientists" by Scientific American--Storm World, and Unscientific America: How Scientific Illiteracy Threatens Our Future, co-authored by Sheril Kirshenbaum. They also write "The Intersection" blog together for Discover blogs.For a longer bio and contact information, see here.


See More

Collapse bottom bar