Anthony Watts says that the number of explanations for the ”pause’ in global warming has now reached 52, one for each week in the year. I used inverted commas to signify that no one knows whether it is a pause or simply a cessation, and no one knows how long it will last. There are two main views: one is that the warming will (soon) resume with a vengeance, and the other is that we’re in for a decade or two of cooling. I think we are allowed to take our pick, and in fact my pick is that I don’t know, and wait to see. Nature will tell us all in time.
Well, it is claimed that excuse #53 has arrived, and in The Conversation, a website about which I feel less than gruntled, given that while it is about universities and proclaims academic rigour it seems to me to be largely an excuse for academics to portray their opinions as science. The piece in question is by a British academic, Professor Rob MacKenzie of the University of Birmingham, and is a bit of a puzzle. You can read it here.
My guess is that it was originally written for a British newspaper or website, because its main focus is on the recent trenchant speech by Owen Paterson, the former UK Secretary of State for Environment, about ‘the Green blob’ — the network of NGOs, sympathetic politicians, businesses and believers who keep the UK in a more and more desperate and increasingly futile search for alternative energy sources that will replace coal-fired and nuclear power stations, at great expense to the British taxpayer.
The headline is probably not the author’s, and it proclaims Climate change: it’s only human to exaggerate, but science itself does not. What does that mean, you ask. The author goes on to explain: To exaggerate is human, and scientists are human. Exaggeration and the complementary art of simplification are the basic rhetorical tools of human intercourse. So yes, scientists do exaggerate. So do politicians, perhaps even when, as the UK’s former environment secretary Owen Paterson did, they claim that climate change forecasts are “widely exaggerated”.
What exactly did Paterson say? The sentence in which this phrase occurs goes: I also note that the forecast effects of climate change have been consistently and widely exaggerated thus far. Well, there can hardly be any doubt that the forecasts of catastrophe have been very widely made, from the UN and within and across all continents. Exaggerated? It depends on what you mean, doesn’t it. Paterson gives examples: The stopping of the Gulf Stream, the worsening of hurricanes, the retreat of Antarctic sea ice, the increase of malaria, the claim by UNEP that we would see 50m climate refugees before now – these were all predictions that proved wrong. I could add a few of Professor Flannery’s forecasts made about Australia.
As it happens, he is absolutely right about the UNEP claim, made in 2005 for 2010. It has disappeared from the UNEP website, but available everywhere by just searching for ‘UNEP fifty million climate refugees’. Paterson’s speech is worth reading — well-written, accessible and a summary of what I would think is the current sceptical wisdom of the possibility of replacing fossil fuels by alternative energy sources anywhere in the next fifty years.
Back to Professor Mackenzie. He argues that when scientists become advocates they are as prone as anyone else, especially politicians, to use whatever rhetorical devices they have to win an argument. I would agree with that. But that isn’t the case, he says, when scientists use their own very special form of mass media, the peer-reviewed literature. There they are cautious. Yes, I would agree again, but the PR departments of their universities are always looking for a good story, and are prone to sex up the implications of an abstract in their media release, in the hope that a major newspaper will pick it up.
MacKenzie argues that another worry is ‘simplification’, which occurs when scientists have to talk to a lay audience. He’s right there, too. And then he sums it up this way: Scientists, wandering unwarily into the realm of advocacy, may be guilty of taking the results out of context, as may be activists and politicians, but it is not the science itself that is “widely exaggerated”… As a taxpayer I would like to believe that physical and computer models provide evidence to politicians who use it to assess the strength of the arguments of the various advocacy groups.
Speaking as another taxpayer I would hope that our politicians do not ever accept that physical and computer models provide ‘evidence’. Model runs are model runs. Evidence comes from observation and experimentation. His last paragraph is a shaft directly aimed at Paterson: I do hope, though, that claims of scientific exaggeration are seen for what they are: advocacy targeted not just at winning the rhetorical argument but also aimed, rather cynically, at undermining the evidence.
I said this article is a puzzle. In my view it is self-contradictory in parts. Does it rank as excuse #53 for the pause in global warming? I don’t think so, though I can see how you could make the case. Professor Mackenzie, for all his wisdom about rhetoric and peer review, turns out, in that last paragraph, to be yet another believer in AGW. Of what evidence is he thinking? UK energy policy is not based on evidence, but on models, alarmism and belief. If it were based on evidence it would have been altered ages ago.
Even though I disagree with Mackenzie’s belief in models, I congratulate him for engaging with his critics in the comments section of “The Conversation”. Unlike a certain psychology academic with his hit and run articles.
dlb
What do you mean when you say
“…disagree with Mackenzie’s belief in models,”
A model is just an explanation of how people understand the world. I presume you must have your own models that you use to make sense of the world.
This model Versus data argument is a bit silly. You obviously need both to make sense of the world
Hi David
Yes, I have “models”, but often I call them analogies, to help explain something, to myself as well as to others. The relationship between models and data is indeed not one of contention, provided the model is being reasonably supported by the data. Where it is not, the data has to be pre-eminent, and the usefulness of the model adjusted accordingly. It needs to be a very good model, proven over some period, for us to rely on its forecasts. Some of our weather models now are really very good, but only for a week or so at the most, I would have thought. But climate models have not shown a good track record, as shown in the SPM of AR5, as well as elsewhere.
Today I was passed a very interesting article on the climate models. I learned a lot. You may be interested, but perhaps you are across all of the issues discussed. Anyway, here is the URL:
http://inference-review.com/article/physical-theories-and-computer-simulations-in-climate-science
I think you’ll find the author is well-credentialled.
We all have models and some are pretty poor. The biblical genesis story seemed to be a valid model for many centuries till evidence proved otherwise. To clarify I have low confidence in modelling complex natural phenomenon such as climate which has too many unknown knowns (e.g. clouds) and most likely unknown unknowns. There is also the phenomenon of drift which render the more familiar weather models useless after a week.
There are two fundamental problems with climate models today, even the most elaborate ones. First, the models are incomplete. This is their cardinal sin. They simply do not incorporate all relevant physics, chemistry and biology. Second, they lack resolution to incorporate all relevant physics, chemistry and biology.
What they do instead is to kludge. For example, sub-grid physics is kludged with various artificial devices, such as Smagorinsky Closure and similar. These devices, used in, for example, aerodynamic simulations, can introduce subtle effects, such as the violation of the second principle of thermodynamics, which whereas invisible in, say, aerofoil simulations, can produce completely skewed, unphysical results in climate simulations that are meant to reproduce decades, even centuries of ocean-atmosphere development. See. e.g., doi:10.1002/qj.2404, for discussion.
Little wonder model based deliverables in climate discussion are so shoddy and disagree so strongly with observations. It is not only with the predicted global temperatures that they fail. They fail to model monsoons, ENSO and many other atmospheric and ocean features that are crucial to our understanding of climate.
Here are some well known failures of this approach:
1. They fail to reconstruct climate change in the past 6,000 years, doi:10.5194/cp-9-1807-2013
2. They produce contradictory projections regarding tropical climate of Australia, doi:10.1029/2011JD017365
3. Their projections are inconsistent with past warming, doi:10.1088/1748-9326/8/1/014024
4. They fail to model correctly atmospheric circulation,
doi:10.1038/ngeo2253
5. They fail to account for climate cycles,
doi:10.1016/j.earscirev.2013.08.008
6. They fail to account correctly for convective mixing and tropical circulation, doi:10.1002/qj.2450
7. They falsely predicted that the Antarctic ice cover would shrink more than the Arctic one, doi:10.1016/j.quascirev.2013.03.011
8. They can’t reproduce observed variations in ground level radiation, doi:10.1002/jgrd.50426
9. They fail the hindcast test, doi:10.1007/s00382-013-1761-5
10. They fail to forecast correctly regional climate changes, doi:10.1088/1748-9326/8/2/024018
In summary, this methodology is so deeply flawed, its deliverables so at odds with observations, it hardly deserves to be called “science.” The ocean-atmosphere-biosphere system of Earth is too complex, too detailed and too large,. to be amenable to this approach at present. People who do this, frankly, waste their time and tax payers’ money.
Don
I saw this comment by Brian Cox on QandA. I thought he was very good. It was a politician free night. Quite good. Have a look if you have not had a chance
http://www.abc.net.au/tv/qanda/txt/s4088125.htm
BRIAN COX: I think there is a misunderstanding as to what scientific statements are. So there is a very simple question, of course, with climate change. If you
consider putting this amount of greenhouse gas into the atmosphere, Co2 and
other gas, then does the climate react and, if so, how does it react? That’s
obviously a sensible question to ask. The only way you can answer that is to
make measurements and to model the climate as best you can. It’s the best you can do. What else can you do? You could read tea leaves and do tarot cards. I don’t know, guess. Well, you can’t do anything else. So that’s what we do. That’s
science. Now, the predictions come back and they have large errors on them. I’m
not sure what the current range is that we carry on. It’s something like 1.7 to
4.5 degree temperature rise by 2100, I think, depending on the assumptions you
make. So there’s a big range of errors. But the point is, if you understand
that what a scientific statement is then you can’t argue with that prediction.
There is nothing else you can do. Science is always preliminary. So it’s not to
be seen as some – we’re not to be seen as a priesthood, which I think sometimes
we are and maybe that’s a problem that we have in communicating that, you know, I’m a scientist and I’m telling you that this is going to happen if you carry
on doing that, therefore you should do this. That’s not the way to do it. The
way to do it…
TONY JONES: Well, you know, the sceptics invoke Galileo though. They say that there is always the chance that everyone else is wrong and I’m right.
BRIAN COX: Well, of course there is. Science is always…
TONY JONES: Galileo proves it.
BRIAN COX: Science is never right. The power of science is it’s the only human system of thought I can think of that accepts its own fallibility. The great Jacob
Bronowski, in The Ascent of Man, said that. He said that this is – it’s the
expression of our humanity because it’s the one discipline that understand its
own fallibility, just like we are fallible. But that’s the point. The point is
that science is the best you can do at any given point. So you’ve got to act on
it. Now, the policy, the policy actions are a different matter. They’re for – I
think that’s where the politicians come in and the democracy comes in. So, but
I get very frustrated, as I’m sure you do by asking the question, when people
attack the science and the modelling in order to disagree with a policy action.
I think that’s where the problem lies.
I thought Cox comes undone with this statement “science is the best you can do at any given point. So you’ve got to act on it.”
As far as I am aware the IPCC climate models have been in use for 25 years and have proved to be flawed. I would think the science is totally different to 18 years ago and no action is required.
Brian Cox… the latest starlet. Note that Brian Cox is an experimental elementary particle physicists. This discipline is as removed from climate science as bridge design. People who work on elementary particles really know very little about geophysics and even less about atmospheric physics, ocean physics, cloud physics and chemistry and so on. Neither are they fully cognizant of the limitations of numerical climate models, the various types of errors that creep into the models from numerical approximations, round-ups, and, of course, incomplete physics, chemistry and biology, and the impact that this incompleteness has on the models.
Experimental elementary particles physics, in comparison, is really exceedingly simple.
The fundamental question of climate science and climate politics is to what extent we can trust climate models and their predictions and how good a justification do they provide for draconian economic and environmental policies, the inevitable outcome of which is going to be massive impoverishment and degradation of the whole society, compounded by severe restrictions imposed on freedom and democracy, perhaps even to the extent of their total elimination. The reason why the question focuses on climate models is because there is no real observational evidence of anything being wrong with climate and the ocean. It is all based solely on the models.
If the models are wrong, then so is all else in climate politics.
It’s not a pause. It’s a maximum. I’ve made a number of bets on this and I’m prepared to pay for a dinner at a good restaurant if this does not become obvious by 2030. What good restaurants there are in Canberra nowadays? In my days “Banana Republic” was good (and new). I wonder if it’s still there.
So you are arguing for a temperature drop by 2030. You are on. 🙂
What are the parameters of the bet ?
I hope I lose.
The loser pays for the dinner, its date and venue to be determined.
Which data set?
Should we use a 5 year moving average?
Do you expect to be alive in 2030.?
It has to be a truly global, properly calibrated data set, so it has to be satellite. It’s the only such.
That’s an interesting thought it’s a maximum. Further down you comment about climate models from my point of view it is simpler than that. I have nowhere near the science background that you have but I am an analyst/programmer with a formal education in it. From a software engineering view these things are all wrong. The are ample reasons that the supporters would want to defend them to the hilt. One of the most convincing things for me would be a specification of requirements. This should include imports outputs and how it was going to be achieved computationally. In computer parlance seems to me from all that I can find they have use the cowboy approach! That is you have a rough idea so let’s code. Inevitably this approach produces garbage. In science I thought you started with an hypothesis and then you test it. In the upsidedown world of climate science they had an hypothesis and seem to have proved it by creating virtual reality models with computers.
“>>> …they have use the cowboy approach! That is you have a rough idea so let’s code. Inevitably this approach produces garbage. <<<"
Indeed. This is exactly how they go about their business. I worked with some of these people at some of the leading laboratories and had an opportunity to inspect their codes. The codes were almost unreadable and there were fundamental numerical errors in some of them, e.g., violation of the numerical stability criterion, because of certain procedures they implemented in their computations, to make the code run faster!!! In this particular case, they had the advantage of being able to compare their codes' predictions against laboratory data and saw, first hand, the divergence, which they couldn't explain… so… I showed them where the problem was and demonstrated how this computations should have been implemented. Well, it was slow, but it was also … correct. They were rather upset about it.
They didn't retract any of the papers published.