Monday
Apr012013
by 
Josh
Josh Introducing 'The Scythe' - Josh 213
Apr 1, 2013 Hockey Stick Josh Apr 1, 2013 Hockey Stick Josh 
Buy
Books
Click images for more details
Currently discussing
-
Jul 11 - logiicmount
on
Seeking Help With Dissertation -
Jul 10 - logiicmount
on
Launched my very own online store.
Links
A few sites I've stumbled across recently....
-
Science blogs
-
Political blogs
-
Other blogs
Subscribe
Copyright © 2006-2013, Bishop Hill All rights reserved.
Reader Comments (84)
Powerful cartoon but an important correction, Josh: others named the original Marcott graph the Scythe - seeking to dramatise the almost upright blade as a killer. Eg Susan Anderson here. But it didn't start with her. See Espen using the term on 18th March, which was common currency by then in warmist circles, for an even worse combination frankenchart than the original.
I fully agree that this horrible metaphor should now be applied not to people liable to die from climate disaster but to all the terrible practices that have led to fear falling on innocents so unnecessarily.
BUt but but...come on...dont you know the seriously flawed stuff is all McIntyres work because he had the temerity to question the religion of Mann Made Global Warming (tm).
Funny though how she cant actually point out the flaws in McIntyres work and instead just goes for ad homs and flat out lies...but what was even funnier is how everyone piled in to her :)
Secondly, does one get the felling that Marcott has been hung out to dry by "The Team"? Even a blind man must realise just how bad their science is? Further more, how many more scientists need to be hung out to dry and made to look like fools before these so called climate scientists start turning on the members of "The Team"???
Regards
Mailman
"I fully agree that this horrible metaphor should now be applied not to people liable to die from climate disaster but to all the terrible practices that have led to fear falling on innocents so unnecessarily." --Richard Drake
Yes, except that innocents have already died in the UK of hypothermia due to unnecessarily escalating energy costs. Those deaths will not disappear and could be just the start.
Yes to both comments. The 'settled science' has never looked so stupid so widely, Mailman. I fully expect Marcott to be hung out to dry. But what about Mann's firm backing? Reversing out of all that isn't going to be pretty!
And yes, jorgekafkazar, the deaths of the fuel poor have already begun. We should be totally focused on that. But the destruction of bad science can only help.
Richard (6:55 PM), I'm not sure you're right.
Susan Anderson wrote "The continuously scything of attacks on credible, widely-held expertise in the many fields working to understand climate are intended to substitute opinion for fact." In other words it's skeptics who are doing the scything.
Interesting btw that Steve M censored our recent exchange. Perhaps it's more widely accepted than we realised that Susan has earned nepotal protected species rights.
Perhaps Marcott et al will now initiate another blaze of publicity and we can look forward to their cheerleaders giving the Scythe maximum coverage. CO2 will freeze before that happens.
Perhaps there are a few key people who can use this sorry tale in a positive way to inform the wider public. The warmists are providing us with wonderful case studies. We should not waste them.
OT: Simon, I don't think anything can be read into Steve Mc's deletion of comments about Susan Anderson except that he reserves the right to enforce a fairly strict standard of what is on or off topic to each thread. He acknowledges that he does not want to devote the time to moderation of the blog to eliminate all comments that he regards as OT (or as "piling on" etc.), so there is some inconsistency as other OT comments survive, but I do think his moderation standard as I have observed it in action has only to do with trying to maintain a focus upon what he regards as the main topic(s) of a thread.
As one who can be chatty and OT (though I try to control it), I have been 'moderated' a number of times by the CA Zamboni, always with good reason, I've thought....
Simon: in my opinion we're both right. Let me try to explain.
I've been mulling over the use of the term "the scythe" for Marcott for around two weeks. I thought it was very nasty indeed - an attempt first to claim that millions were bound to die because of the uptick, the blade (as I took the uptick to be, unlike a normal scythe) and then of course to say that we sceptics were to blame for this 'holocaust' by scything down the wonderful, pure science of the Team and persuading politicos to do nothing.
I don't have very nice thoughts about some of these people. This is what I was expecting.
But on Easter Day or shortly after, GMT, I wrote down "O Scythe, where is thy sting." I won't give you the full chapter and verse, as they say. But I did feel that this very evil piece of propaganda would now fall out of the enemy's hand and into ours. But we wouldn't be using it against people but against the lies they've been told, using it most of all on behalf of the defenceless.
So you're totally right about phase A, as Susan Anderson (bless her cotton socks) demonstrates, albeit with some hesitation, which is good. The real haters live further below the surface.
But I'm also right to have written on CA early London time:
Their science is so bad that perhaps it's fair to say they've self-scythed. And lo, we are both right :)
Skiphil: I genuinely rejoice whenever I am zambonied by Steve. I think many online fora have so much to learn from his brilliant achievement with CA, in the most adverse circumstances possible. But I've been accused of saying too many nice things about the guy recently so I must now assume an air of indifference :)
"But I did feel that this very evil piece of propaganda would now fall out of the enemy's hand and into ours. But we wouldn't be using it against people but against the lies they've been told, using it most of all on behalf of the defenceless." Richard, I agree wholeheartedly. (As an aside do please fill me in on what "zamboni" means).
Shock horror, the sainted Steve McI deletes posts? Can I be reading that right? Where's Geoff Chambers - he'll be whining about censorship on Climate Audit, I'm sure of it!
Oh and by the way, if you re-plot the "Scythe" without the uptick and then plot the instrumental record on the end you get a sort of hockey stick. So what changed?
Josh ... wonderful as always!
Over his FB, Mann credited the term to this Daily Kos post.
OT: Bit, as far as comparing moderation policies goes, the huge difference between CA and what the NLP is doing (or what is done at Real Climate or Skeptical Science) is that that McIntyre goes out of his way to retain CRITCAL comments even when they are (often) not on topic or not civil. He bends over backwards to engage critics in discussion. NLP clearly is censoring civil on-topic comments which appeal to reasoned arguments, solely because the blog moderators don't like the views expressed. That is their "right" as any blog/website can do it as they choose, but it is heavy-handed and hostile to reasoned debate. They act as though they want only the cheering section.
McIntyre, whatever you may think of him, energetically welcomes his critics and preserves virtually every critical comment -- he "Zambonis" some "friends" of the blog who go OT or are overly repetitive or emotional, vituperative, etc. You can find many, many threads there where critics engage Steve Mc at great length, often with more relaxed standards of what is "OT" or "civil" etc. than what is allowed for the friendly commentators.
The key skill of the Global Warming Industry is its ability to juggle memes and slogans and ideas rather than the validity of its flawed hypothesis.
"The Scythe" can be a rallying cry for those who point out that global temperatures are flat-to-declining. Let's all use it as shorthand for the busted and bankrupt Global Warming hoax.
There seem to be a lot of those defending Marcott, such as BB, that do not seem to realise the implications of the resolution difference between the reconstruction and the instrumental data. Basically, if you have a high resolution, such as the instrumental record, you see all the short term noise (the spiky bits) - you have data points down to day by day records. The long term reconstructions have a resolution of hundreds of years (smoothed to 300yr if I recall). That means all the short term spikes are lost. Ergo, if you plot the instrumental data at the same resolution, the last hundred years or so would be represented by a single data point - no uptick; or if you had a fully resolved past record, it would appear just as noisy and have upticks (and downticks) all over the place.
It's an apples and oranges comparison. Try to splice them, and you end up not with a scientific paper, but a fruit salad.
BitBucket - you come across as more bitter and twisted than ever.
What changed? The temporal resolution, by 2 - 3 orders of magnitude.
You could have worked that out yourself.
The pleasure of having 99% of my CA comment attempts deleted over the years has been all mine.
I deserve a figurative knuckle-rapping from Teach to keep my Badness from distracting everyone.
He was right to do it.
Andrew
Listen, I don't give a flying frack if there is a difference between CA's 'moderation' of comments he doesn't like and NLP getting fed up with a barrage of sceptics' comments. They are both private blogs and can do what they please. If you don't like it, start your own blog and empty your mind there. Geoff Chambers calls NLP's action 'censorship' - is it or isn't it?
Bitter and twisted? Get a new tagline Martin, that one's stale.
Temporal resolution, my foot. Let's say we plot the Marcott curve up to the 1850 reconstruction end point and then add one point for the average temperature anomaly between then and now. Does it go significantly up or down? Since you are 'sceptics' I'll give you three guesses, that should cover all options.
The resolution end point is not 1850. Read McIntyres post on proxy 23. The last 3 points were
'A sample from 3 cm depth in the MC core showed a bomb spike. The high sedimentation rate (average 31 cm/ka) over the last 7000 years permits analysis at multidecade resolution with an average sample spacing of ~32 years.'
A bomb spike records the signal from airburst/surface detonated nuclear tests. There were no nuclear tests in or before 1850. Furthermore 'Despite this evidence for modern sediments, Marcott et al blanked out the top three measurements as shown below' and follows with a graph showing the modern proxy determination as a red dot of even lower proxy calculated temperature.
http://climateaudit.org/2013/03/17/hiding-the-decline-the-md01-2421-splice/
This graph, the truncation of either end, on the modern end because the proxy goes the 'incorrect' way, and the earlier glacial end because it gets people thinking about real 'catastrophic climate change' encapsulates the mischief that we are all so wearily used to and disgusted by.
Bitbucket-
"Let's say we plot the Marcott curve up to the 1850 reconstruction end point and then add one point for the average temperature anomaly between then and now."
Let's not. Instead, let's follow the paper's argument, which claims that 0% of the temperature variation is recovered for periods less than 300 years. Indeed, only 90% of the temperature variation is recovered over a 2000 year period. Average the thermometers from 1850 to 2012, and multiply by 162/2000 to get an equivalent 2000 year point.
Does it go significantly up or down? Can you even tell?
What a stupid game.
Pharos, how many proxies were there, 73? And you are worried about number 23. Nice one, I bet it is very special! So it is a mud core is it? Ever seen a mud core? No neither have I. But I can guess what the core-top of such a sample looks like and it wont be like a stick of Brighton rock; more like what goes down the loo after a night on the curry. So you measure 3cm into that sludge starting from some arbitrary 'top' and you want to draw your great conclusion from that? Genius!
Chrisy, I have no idea what that means, but I'm assuming you mean that short term variation does not come through. Well Tamino does a simple average of the data and gets a noisy curve very much the same as the smoothed Marcott curves - Tamino's basic plot is noisy at a resolution of much less than 300 years. And remember that the Marcott processing is designed to combine 73 noisy data-sets - you'd have to be stupid to apply the same process to a single relatively clean instrumental data-set. So my suggestion stands.
BitBucket
I said you come across as more bitter and twisted than ever.
if you re-plot the "Scythe" without the uptick and then plot the instrumental record on the end you get a sort of hockey stick. So what changed?
What changed? The temporal resolution, by 2 - 3 orders of magnitude.
You could have worked that out yourself.
Apr 1, 2013 at 9:31 PM | Martin A
Marcott et al themselves regard the 20th century part of their data as less reliable than the rest, and have said so repeatedly.
Why are you so keen to have them use poor quality data when the instrument record provides data with improved "temporal resolution, by 2 - 3 orders of magnitude"?
Why, too, are you all concentrating on a peripheral part of their research, when the core is on the whole Holocene?
Why has Bishop Hill devoted at least four (4) posts and two (2) cartoons to this paper, along with similar efforts from McIntyre and Watts, when it says nothing much about current climate change at all? :-)
Some would say that Marcott et al. brought the attention on themselves by attempting deceit - this is their abstract from Science, after all:
Clearly the claim that 'Current global temperatures of the past decade have not yet exceeded peak interglacial values but are warmer than during ~75% of the Holocene temperature history.' has been shown to have been fabricated. Anybody who defends this work shares the credibility of the authors. i.e. nil. The journal hasn't done too well either.
"when it says nothing much about current climate change at all? :-)"
In fact, it says nothing much about anything at all, and does it so ineptly that it's the laughing stock of climatology. Roger Pielke jnr sums it up beautifully.
And watching the 'Team' running around in ever decreasing circles trying to defend it (and Science for publishing it) is the best laugh I've had in a very long time.
I thought the blade of the scythe was in the upwards tick part, the handle being the "looong slope downwards"
But there you go, as a meme travels..
it says nothing much about anything at all, and does it so ineptly that it's the laughing stock of climatology.
Apr 2, 2013 at 12:04 AM | Cumbrian Lad
It gives a useful summation of the current data on the Holocene and puts the instrument record into perspective. In 100 years the instrument record shows a temperature range of similar magnitude to that of
the entire Holocene.
None of the sceptics have demonstrated previous temperature changes of that scale and timescale. The inability to supply precedents badly weakens your case that the 20th century warming is natural.
I have watched the response here and on other sceptic websites and see you running around like ants. I dont see any climatologists laughing either.
BB
'Ever seen a mud core? No neither have I.'
Sorry to disappoint you, not only seen but logged them and done sedimentological and palaeo analysis on them in the lab.
"Why are you so keen to have them use poor quality data when the instrument record provides data with improved "temporal resolution, by 2 - 3 orders of magnitude"?"
We're happy for them to use both, but not on the same graph, where the different resolution gives a totally false impression to people with low critical skill.
Stop digging, Marcott has.
Reading off Marcott et al's graph, using mean values, the anomaly temperature was -0.2C 11,000 years ago. It peaked at 0.45C at 7000BP and reached a minimum of -0.4C at 200BP.
The temperature range for that 11,000 year period is 0.85C and the typical rate of change is 0.13C per millennium.
The instrument record (GISS data) shows a change from -0.45C in 1910 to +0.66C in 2010.
That is an increase of 1.11C. Over that 100 year period the rate of change has averaged 11.1C per millennium.
This is probably the root of the debate. Despite much noise and smoke I have seen no reliable demonstration from a sceptic that anything comparable to the modern changes has taken place during the Holocene.
Remember that under your own rules, the interpretation of the null hypothesis used on this site, you must assume that the short term variations you postulate do not exist until you can demonstrate their existance.
Brilliant, as usual, Josh.
Reminds me of the 'Dark Riders' in L o t Rings, watched recently with my small sons, 26 and 21!
Will we Hobbits prevail?.
Not if we only talk to each other.
Time to engage with the MSM with letters or a petition.
C'mon Bish, you have the skills and an audience to do this.
Pharos, that is no disappointment; quite the opposite. You are ideally placed to tell all on the Hill what we don't know.
So what is the likely appearance of the core top of core-23? And how does one determine where the 'top' is? Is it like a Brighton rock or like the sludge at the bottom of your garden pond? You quote an average core density (31 cm/ka) but how does that vary over the core? Is the density of the core top the same as that at the core bottom or is it very different? Is 3cm or half a pinkie from the core top a very meaningful measure? What is a better measure of the age of samples along the core - its radio-carbon date or its position relative to the 'top'. So many questions and you are the one with the expert experience to answer them.
Bitbucket-
"Chrisy, I have no idea what that means, but I'm assuming you mean that short term variation does not come through."
Actually, you should read the paper's supplemental material, section 9, pages 23 - 26. If you can't be bothered to read, then just optically look at figure S17.
The authors of the paper demonstrate that their temperature anomaly response is as follows for the signal period considered:
1 year- 0.0
(0% of the amplitude of a temperature anomaly with a 1 year period will show up in the construction).
2 years- 0.0
5 years- 0.0
10 years- 0.0
20 years- 0.0
50 years- 0.0
100 years- 0.0
200 years- 0.0
300 years- 0.0
500 years- 0.1
750 years- 0.3
1000 years- 0.5
1500 years- 0.8
2000 years- 0.85
3000 years- 0.9
5000 years- 0.98
10,000 years- 1.0
In other words, for a 150 year integration period as you suggest, the authors claim that their proxy construction is able to accurately capture about 0% of the temperature fluctuations.
The construction only captures about 10% of the amplitude of a 500 year long temperature swing. This length of time is comparable to the MWP.
Therefore, comparing 20th century trends in Temp or d(Temp)/d(time) with the proxy results from this paper (as has been done by the climateer glitterati) is specious drivel.
Entropicman and Bitbucket,
Do you actually realize how damn stupid and ignorant you both appear to be?
Mike:
No, they don't. No self awareness whatsoever.
Entropic man Apr 2, 2013 at 12:39 AM |
"I dont see any climatologists laughing either."
No? They've been laughing all the way to the bank - and Tahiti - for about 25 years now.
Chrisy, so they made a low-pass filter. Big deal, that is how you separate low frequencies (the trend) from high ones that are different in each proxy (the noise). But the thing about the instrumental record is that it has almost no high frequency noise; it is all signal. You can tack it on the end and there is no conflict. The up-tick is real. The problem is that it is impossible to tell from the graphs whether there were up- or down-ticks at other places along the trend curve. Maybe it is there in the data if processed differently, I don't know. I'm not aware that Marcott et al claim that there were no ticks along the way. Do they?
Old Mike, not nearly as stupid as someone who has a 30-year stockpile of incandescent light bulbs...
chris y - the words "it doesn't matter" have a long and proud history in climate science. To Entropic man the fact that any short term variation (positive or negative) over 150 years is never likely to be observed in a chronology with a resolution of at best 300 years just, well just ....
"doesn't matter"
New Horror movie: The Scythe, The Scythe!!
btw, for anyone who has not followed the latest developments with the Marcott et al. study, there are some interesting items collected at the Dot Earth/NY Times blog. Also quite a dust up in the comments section, where a few of us are striving to bring some needed balance and perspective to all the "Steve McIntyre is the spawn of Satan" types there:
updates on Marcott et al. (2013) at Dot Earth/NY Times blog (Revkin)
The New Left Project has nothing on some of these comments (for insularity), except that the NLP seems to attract a more intellectual type.
ptw:
That was always how I interpreted it too. Josh attributed something else to my words, given the shape of the blade of a real scythe. But, whatever the shape of the blade, the metaphor is now being used in exactly the way I wanted. With the brilliant outing of Marcott's bad science by Steve, Jean S and other, spurious scares based on specious drivel - thanks chris y - must themselves take a fall.
Simon Anthony:
Another fine metaphor, used in olden days on Climate Audit, given that the host is based in Toronto. I'll let you work out the rest.
Skiphil - "Steve McIntyre is the spawn of Satan".
And to complete the quinella is he "what Lysenko spawned" as well?
Skiphil: fantastic work, with theduke and others, on the NYT comments. And I liked this from Mosher:
Mosh almost certainly judges the risk to be greater than I do. But the logic's inescapable: if the threat is real it doesn't need quackery like the Marcott uptick to promote it. That's totally counter-productive - a self-scythe. They have got to learn.
Jean S: So The Daily Kos originated the term on 9th March. I knew it wasn't me. Thanks!
news flash for anyone who can try to run down estimates of insolation over the period of the Marcott et al. (2013) study!
New comment at CA, and I looked at the linked reference, it definitely only shows years 1500 - 2000 in Figure 4, with rising not falling curve, so there may be some mystery about where the claim for 11,500 years of decreasing insolation comes from, is it accurate claim but simply wrong reference given, etc. Or what is going on and what does it mean for the science of the Marcott study?
seriously mistaken reference and claim about insolation over 11,500 years
[emphasis added]
Richard (Apr 2 4;14 AM) Got it. Thanks. Interestingly, the term is also quite apposite when describing Marcott's smoothed historical record. And btw I think you're hijacking the word "scythe" (Apr 1 8:04 PM). As I remember it the text is "O death where is thy sting, where grave thy victory?".
There's cherry-picking and then again there's cherry-picking.
And then there is a simple abuse of the English language.
Just exactly how is "Over that 100 year period the rate of change has averaged 11.1C per millennium" supposed to be in any way meaningful. Extrapolating 100 years forward to 1000 may be very clever but is about as sensible as my saying that the rate of change over the last three hours has averaged 24.8C per day.
Do at least try to talk sense.
Mike, exactly my point when accused of driving at over 30 mph. Where are the thirty miles, where the hour?
EM, this is indeed the root of the problem. What you seem to be refusing to understand is that, if for the sake of argument there had been some unknown event that occured 8,000 years ago, that resulted in temperatures rising by 1C over 50 years, and faded away as rapidly, then the spike on a high-res graph would be more dramatic than the recent AGW spike, but it would be completely lost on the Marcott scale.
This is not in any way to imply that such an event occured. But you cannot use Marcott data to show that it didn't occur. You cannot use Marcott data to show that the recent changes are unique in the Holocene.
Why cannot you accept this rather simple point?
Thanks steveta. Picking up on just this piece
is it not fair to say that the only thing that either sceptic or alarmist can demonstrate totally reliably is that the human race started to use thermometers systematically, across the world, around 1850? We know that date but we are still struggling with the intellectual consequences. Incomplete knowledge is for some a fearful thing.
So a few surprises this morning. Firstly there's no answer from Pharos, the man who has seen cores. He wants to tell us, like McI, that the top 3cm of a core can be used to identify subterfuge on the part of Marcott et al, and yet wont tell us what such a core looks like. My guess is that you can't tell diddly-squat from the sludge at the top of a core. But what do I know?
Secondly Mike Jackson and Simon Abingdon (9:11/9:16) think it laughable to measure as speed of 30mph over a distance of less than 30 miles and a time of less than and hour. I don't know what one can say about such silliness. I know they are not as stupid as that sounds, so maybe it was just morning confusion.
Thirdly, my last post from last night is not there. So here it is:
Chrisy, so they made a low-pass filter. Big deal, that is how to separate low frequencies (the trend) from high ones. They need to do that because the high frequencies are different in each proxy (they are noise). But the thing about the instrumental record is that it has almost no noise; it is all signal. You can tack it on the end and there is no conflict. The up-tick is real. The problem is that it is impossible to tell from the graphs whether there were up- or down-ticks at other places along the trend curve. But I'm not aware that Marcott et al claim that there were no ticks along the way. Do they?
Old Mike, not nearly as stupid as someone who has a 30-year stockpile of incandescent light bulbs...