Bishop Hill The debate at the FST
Jul 2, 2014 Climate: Statistics Climate: Surface Walport A report of the recent climate change discussion at the Foundation for Science and Technology has been published here. Audio of the main speakers is available from the FST's website.
Featuring Mark Walport, Jim Skea, Peter Lilley and David Davies, the subject was "What is the right level of response to anthropogenic induced climate change?". From the report of proceedings, little new ground was broken. I was, however, interested to learn from Walport that it is "clear" that climate change is happening and that its impacts are already evident, a position of delicious imprecision: I imagine we are supposed to infer that he means manmade climate change, but of course manmade climate change is not "clear". As I have mentioned previously, I have put it to Walport that we are unable to demonstrate a statistically significant change in surface temperatures because of the difficulty in defining a statistical model that would describe the normal behaviour of surface temperatures, a claim that seems to have the support of the Met Office. I don't know of any other metric in which a statistically significant change has been demonstrated. Walport did not dispute my position on surface temperatures but suggested that seeing many observational metrics moving together led to a conclusion that manmade global warming was upon us.
This may be the case, but I wonder if there is a robust statistical analysis of to support Walport's position. Perhaps a letter is in order.
(Please could we avoid comments that are simply venting about Walport - stick to the issues please.)
Reader Comments (46)
Walport is merely repeating the same old mantra about climate change. He cannot be specific, nor can he provide any evidence as there is none. If there were any we would have been shown it. These "warmists" believe that you must repeat the same line ad nauseum. Now temperature is at a stand still it is all they have got.
What is the right level of response to anthropogenic induced climate change?
The proper response might be: what anthropogenic induced climate change (as such a phenomenon doesn't exist). But, hey who cares, it's the computer models that count and the climate itself in the real world is the garbage. Anyway, I suspect the irony might be lost on the climate crowd and proceedings would still get into gear without pause for thought.
"seeing many observational metrics moving together"
Is an 'observational metric' the same as an observation, or just a computer model in disguise?
If they are so obvious (to Walport, at least) it shouldn't be too difficult for him to identify them - should it?
"we are supposed to infer that he means manmade climate change"
We usually are, but it is rarely stated, as it invites a rather obvious response! To be fair, "anthropogenic" was in the discussion title, but that soon gets forgotten, which is curious from people whose normal MO is to use six words whenever two will do.
Is he the man wearing the parachute in the Admiral Insurance adverts?
I googled "observational metric' and got 501 hits. Using an observational metric myself, I found most attached to climate with the majority of the remainder being cosmological. One can forgive the cosmologists for working with what they can get but clima people have no excuse for not measuring.
Perhaps Walport should stick with the more traditional "anecdotal evidence" to spare his blushes.
[Snip - I've asked people to stick to the issues]
I had a similar experience with a very experienced scientist on Twitter. She (an astronomer with zero climate science knowledge) was 1000% convinced of CAGW and even Tamsin could not move her away from such a strong religious Belief.
There are many scientists who think their specialty is a mess but the rest of science is rock solid. This makes the position of Chief Scientist a very dangerous one, and Walport is just the latest example of why the post should be abolished.
The debate was mistitled. The title should have read: "What is the right level of response to anthropomorphic climate change?".
There, all better.
Was it Monckton who said, vis-a-vis response to climate change: "We have to be brave enough to do nothing."?
The observational metrics that show manmade warming - as defined by the IPCC - were stratospheric cooling, a tropical troposheric hotspot and unnatural ocean warming. The first has been absent since 1995, the second is not apparently there at all in the obs (rightly or wrongly) and the ocean has not even warmed in the top 700m (naturally or otherwise) since we installed a system of measuring it in 2003 - and this is after they added the mandatory warming adjustments. What possible other metrics can he be thinking of? And if there are any, what justification is there for using them instead of the IPCC ones?
That doesn't mean a thing unless you know how many metrics there are, how often they changed in the past and to what degree they changed.
If you have a thousand* different metrics all of which vary on an annual basis then there is a fairly good chance that several of them will experience a statistically significant change purely by chance. Since every change, whether up or down is ascribed to anthropogenic caused climate change getting "proof" is easy.
* A thousand might seem a large number but it isn't. Just for the UK you can high, low or average temperature, precipitation, sunshine, snow for year, season, month or day in England, Scotland, Wales, Northern Ireland, South East, South West etc.
Personally I thought that the report of the debate, if it was a true account, was rather good.
There certainly was not the stereotypical conclusion outlined by the normal environmental NGO's and rather than the 'case is settled we need to do this' it leaves the reader to draw their own conclusion from the evidence presented, as it should be in open debate.
As Lord Beaverbrook points out, the report of the discussion is equivocal:
"But there was disagreement about by how
much global temperature will rise, and how
much is due to anthropogenic activity, and
whether subsidized renewable power
generation, which inevitably increased
electricity prices, was appropriate for the UK.
"There was a strong disagreement between
those who considered we must act now and
should not wait, and those who thought that
we should only adapt and take action when
we knew that temperature rises were taking
place. The latter thought that the UK target
of reducing GHG emissions by 80% from
1990 levels by 2050 was premature."
Peter Lilley's comments were skeptical of the warmist assumption.
Surely, an 'observational metric' is eyeballing what's displayed on a computer screen?
"While it was, perhaps, fair to criticize scientists for not being able to predict how long the current pause in temperature rise would continue, and say that we would only know if it was temporary or not in 50 years, the criticism did not take account of the many other observations and findings that had been made about temperature changes and, importantly, their speed and relationship to emissions."
But what are these "many other observations and findings"? If they would bother to ask that question then they might actually gain a perspective of the sheer level of circular reasoning and confirmation bias present in the climate science community becasue the only answer that is possible is that the believe their models - even though they are demonstrably errant. To do otherwise would be to admit there is no problem and that they should be defunded.
The greatest achievement of anthropogenic induced climate change is to have kept global surface temperatures within an unbelievably narrow range for many hundreds of years - with virtually no change for the last 17 + years. So the right response should be a round of applause.
"What is the right level of response to anthropogenic induced climate change?" is exactly equivalent to the old saw: "Have you stopped beating your wife yet?" - it assumes any response is an acceptance that AGW is real.
As Rick Bradford says, about Monckton's "Be brave enough to do nothing", my favourite response is based on the wonderful (late) Keith Waterhouse: "Don't just do something, stand there!"
"seeing many observational metrics moving together" would only be a reasonable metric itself if they could demonstrate that current weather was markedly different from that seen when circumstances were similar, i.e. the current plateauing of temperatures achieved after a prolonged and deep cold phase. Since as far as I can see that was around 1600 years ago, I don't see that they can which makes such statements entirely meaningless.
http://www.bbc.co.uk/news/science-environment-28089165
Check this out. Now we can have another argument about how you measure Carbon Dioxide.
See them make another Hockey Stick out of Satelite data
Ed Hawkins wrote a blogpost yesterday that attempts to say when the 'signal' breaks out from the 'noise' in the temperature record. But these concepts are defined in an arbitrary way and can't really be clearly defined. He has a paper coming out on this I think. Also of course this says nothing about whether it's anthro.
I was taught at school that you should never predict anything without having all the necessary historical evidence to hand - unless you are a stock broker when forecasting (tipping) can be very lucrative without any evidence at all.
Paul wrote: "Also of course this says nothing about whether it's anthro."
Indeed. The Pause is doing nasty things to the attribution arguments, as well as sensitivity, but the issue is successfully buried by a simple reference to "climate change" without qualifiers. People who argue against comments like Walport's are falling into a rhetorical trap...
"The Pause" is climate change.
extra CO2 release is responsible for 10-20% more plantlife and an increased habitat for the animal kingdom.
This to the chagrin of every selfrespecting vinegar pssing progressive libbie who prefers scourched places like Haiti and Somalia (both "free" and anti western in sufficient measure but always in need for a little "help" for the last 50y)
hunter
"anthropomorphic climate change"
I'm not sure that's right either. Better than 'antropogenic' which is to do with the development of humans, but 'anthropomorphic' is to do with appearance or the impression of human characteristics, usually incorrect. On reflection, perhaps that is the word to use.. :-)
I agree with Mark Walport that it is clear that climate change is happening. What was the last time the climate was not changinng? Is he a creationist?
jamesp,
Thanks for the feedback. "Anthropomorphic" was arrived after careful consideration.
Anthropomorphic visions project human fears and nightmares into the dark night and blue sky and comes back with demons and divine punishments....all from the human mind.
AGW is the modern version in effect of a huge free form Rorschach test.
Our politicical and academic class have looked into the skies of Earth and found excuses for sorry academic work, derelection of duty to prepare for known weather risks, poor critical thinking skills, historical illiteracy and so much more.
Thank you for not snipping my comment at 12:02 PM.
For those who are mystified, see the link below at around 10 seconds.
A dead ringer, I reckon.
http://www.youtube.com/watch?v=qu9z36itZtY
As you say, little new ground was broken. Except in one important respect; namely that this debate took place at all with a balance between the points of view. Set against the BBC prejudice to deny airtime to the views expressed by Peter Lilley or David Davies, this is significant. I am glad to see the Foundation put on this debate and the Chairman asserting its role as a neutral platform.
I thought the points put forward by Peter Lilley and David Davies to be eminently sound.
You all have to learn to stop arguing about "climate change" (and accepting it as it is used by the alarmists--as an empty political phrase), and saying silly things like "climate is always changing". The GLOBAL climate (i.e., the global mean surface temperature) is NOT changing, as my definitive Venus/Earth temperatures comparison proves.
> As I have mentioned previously, I have put it to Walport that we are unable to demonstrate a statistically significant change in surface temperatures because of the difficulty in defining a statistical model that would describe the normal behaviour of surface temperatures, a claim that seems to have the support of the Met Office.
The problem, of course, is to show the relevance of that (and such) claim in the grand scheme of scientific things:
http://andthentheresphysics.wordpress.com/2014/07/02/come-one-andrew-you-can-get-this
Comments and corrections are welcome. Concerns, are also appreciated, as always.
Thank you for your concerns.
There are vast numbers if changes: a few years ago someone came up with a correlation between penguins and wearing dinner jackets.
I would suggest that the reason why CO2 is so popular is that it is relatively easy to measure. Factors which influence climate could be
1. Cosmic radiation.
2. Energy output of the sun.
3. Frequency of output from sun.
4. Variation of spin if earth's axis.
5. Distance from Sun.
It is much easier to use changes in CO2 as the basis of a computer model than items 1 to 5. Surely Walport needs to disprove items 1 to 5 as well as prove CO2 is having an effect.
Changes in climate have occurred on Earth for the last 4.5 B years.
I would suggest that no statistical model is suitable until we have defined the actual temperature record for at least the last 2 M years..
Lilley's talk is pretty good if one excludes some infelicitous statements about climate science.
Perhaps you might invite Walport to comment on the 'observational metric' that Antarctic sea ice has just broken the satellite era anomaly record and that global sea ice area anomaly is currently around +1 million square kilometres.
http://arctic.atmos.uiuc.edu/cryosphere/
That should persuade him to rummage through his lame excuse repository looking for the many remaining 'observational metrics' that are moving together. Since the surface temperature record has also taken a hit recently it looks like there are precious few except perhaps for butterflies moving north to escape the heat.
An observational metric is simply anecdotal evidence. It's a personal impression. That is to say it's worth diddly squat, scientifically. In climate science terms, however, obviously it carries a lot of weight. That's climatology for you.
Walport's comment re observational metrics is very similar to a bad Cricket commentary; an overblown euphemisim for describing a quite unremarkable event, all adding nothing to any listener's knowledge, except to add to the general impression that Walport could 'play a blinder' (if he actually knew what he was talking about), in a similar fashion to one of the luminaries being complimented after being questioned, and saying nothing, during the course of a well-known enquiry.
Willard
I read your blog post. Next time if you want to call it physics you shouldn't assume we understand what is called the Greenhouse effect just because someone heated some gas in a box.
To be physics you need to eliminate convection effects as well as boundary effects (the gas cannot freely expand) plus add in the simple fact that to have an effect it needs to be characterised. Otherwise you are making the Theoretician's mistake - assuming you know everything.
The reason a statistical model would be used for temperature is because the measurements aren't accurate enough to create an anomaly calculation. Hence assumptions are made and a model used. The problem is that it still means it's fiction.
Number one rule of science: don't present theory as truth without caveats.
The comment by Ed Hawkins, Richard Betts, Doug McNeall and others, criticising a paper by Mora et al last year on the time of "climate departure" (relevant to the question of Walport's "clear") is here. It's quite outspoken. "Several methodological oversights contribute to the erroneous uncertainty quantification."
The story is also at Ed's blog and at Climate etc.
Chris Rapley will be cross. What happened to the 'meta-narrative' and 'coherent unified voice' he's calling for?
Remember when willard says "comments are welcome" he actually means "I cannot discuss here since I cannot snip other people's comments at will so please come and waste your time at the physics-challenged blog run by the partial amateur who doesn't understand measurement units and where former bishop-hill trolls psychologically support each other".
In any event it seems we are well on course to avoid 2 degrees of warming just by business as usual. So the correct response is to work on the main problem which is providing enough energy to cope with the expected increasing demand by fossil fuels and/or nuclear power plus any efficiency gains in the gas-power cycle (eg solid oxide fuel cells) until such time as nuclear fusion takes over. If solar cells make a huge breakthrough meantime then great. If geothermal energy can be done cheaper then even better. I hold out little hope for wind or wave energy. Please uk.gov, stop wasting money on feelgood projects, stop paying extorsion money to foreign nuclear companies and instead ramp up our own nuclear capability again. For the same money as HS2, it's possible to build 5 to 10 home-designed and home-built nuclear power stations. Even Walport is big on shale and nuclear I believe so there is at least common ground if we ignore the zealots.
On And then there is physics, Willard(?) said
(my emphasis)The Bish was quoted as saying
"we are unable to demonstrate a statistically significant change in surface temperatures because of the difficulty in defining a statistical model that would describe the normal behaviour of surface temperatures "
Seems to me that what the Bish said is completely in line with the absence of any adequate statistical model for surface temperature.
Yet Willard then continues with stuff like "...So, Andrew, I really do think you can get this. It’s not that tricky. I appreciate that it may be embarrassing to have to admit that you’ve misunderstood something so simple...."
No doubt that sort rudeness to people you disagree with goes down well with The Faithful.
I posted the following at And now there is physics. ("Your comment is awaiting moderation")
____________________________________________________________________________
I think there may be some misunderstanding of a key point here. If you have a statistical model for the process that generates a random signal, you can use this to decide whether what you are observing is:
- simply the signal fluctuating as normal
- the signal is no longer fluctuating as normal because something has changed significantly and what you are now seeing is unlikely to have been generated by a process described by your model.
This is standard stuff in (for example) detecting the seismic signature of a nuclear detonation against background seismic activity. Or the noise of a nuclea submarine against the background of normal undersea noise. When your tests show that something that no longer has the normal statistical characteristics, you can infer that something has changed.
My understanding of Andrew Montford's point is that, in the absence of a statistical model for climate variations in the absence of anthropological effects, you can't say whether the recently observed changes are out of the normal range of statistical variability. I think you agree with him on the nonexistence of adequate statistical models.
From the FST report: A speaker asked how he would know if $10bn was spent over ten years on emission reduction was worthwhile; how could we measure success? The response that ‘many things would not have happened which otherwise might have’ was unlikely to convince the public.
I was the questioner. I did not say $10B, but $10T, a much greater sum, and the answer is therefore even more preposterous given alternative calls on that sum which would have measureable outcomes!
I have written in to have the factor of 1000 restored. One could write off billions associated with a failed IT system for the NHS, but trillions are a quite different matter to bandy about and to commit to (possibly) stopping change.
As I have written elsewhere:
And the problem with such extraordinary levels of spending is that it is hard to discern whether it is worthwhile. What do you actually get for £1 trillion or £10 trillion spent on climate mitigation? The answer is no one knows. If there is catastrophe later, we clearly underspent; if there is no catastrophe, we will never know whether we can claim credit for having averted it. So in either case the money is spent to no known good effect, despite there being competing calls for the money that would at least have measurable positive consequences. That may in fact be the rational reason for past inaction.
wallport = (*)
Most of tracegas plantfood release in the atmosphere has been beneficial for nature.
that's an inconvenient FACT, that is.
> Seems to me that what the Bish said is completely in line with the absence of any adequate statistical model for surface temperature.
Rejecting all models as inadequate is not the same as accepting that no adequate models exists. In the former case, it's a rejection based on an appeal to perfection. In the latter case, it's an acceptance based on common sense.
To claim that both stances are "completely in line" with one another seems to be farfetched. In fact, here's how Richard Muller characterizes the first endeavour as an "exaggerated pedantry for statistical methods":
> What he is saying is that statistical methods are unable to be used to show that there is global warming or cooling or anything else. That is a very strong conclusion, and it reflects, in my mind, his exaggerated pedantry for statistical methods. He can and will criticize every paper published in the past and the future on the same grounds. We might as well give up in our attempts to evaluate global warming until we find a “model” that Keenan will approve — but he offers no help in doing that.
http://neverendingaudit.tumblr.com/post/11763136868
Interestingly, this excerpt comes from an email correspondence that Douglas released. We still don't know if he asked Richard before releasing that correspondence.
I think that same point applies to Martin A's comment.
***
In any case, thank you for your comments, Micky H and Martin A. I'll forward them to AT, who is not me.
First rule of ClimateBall (tm): make sure you identify properly who you're trying to hit.
Michael Kelly (I assume):
You have stated this crucial point more clearly than anyone I've seen. No measurable impact - a recipe for irrationalism in energy policy that quickly translates into arbitrary crony capitalism of the worst kind. Very glad to hear you asked the question.