Once again, the Team has “moved on” so quickly that it takes some care keeping track of their movements. The criticisms in my most recent post apply to the still unpublished Tingley and Huybers 1200-year reconstruction at their website (that it uses Mann’s PC1, a second strip bark foxtail series, Yamal plus a van Engelen series that even the IPCC acknowledged could not be used as a “proxy”). This reconstruction (let’s call it TH2009) is a typical small subset reconstruction (14 series), in which the primary issue is data snooping – the re-use of data sets with known and even stereotyped properties – issues that were raised in my previous post
As I noted in the update to that post, it appears that there is another unpublished Tingley and Huybers submission covering only the past 600 years that isn’t posted at Tingley’s website and that it is this other unpublished unposted submission that is featured by David Appell in this month’s Scientific American. This other reconstruction appears to be the one presented by Tingley and Huybers at the PAGES 2009 conference here. Let’s call this reconstruction Tingley-Huybers 2010. Here is a plot of the 600-year TH2010 reconstruction from the PAGES PPT (reshaped here to facilitate comparison with other sticks.)
Figure 1. TH2010 reconstruction (north of 45N).
The TH2010 Network
The TH 2010 network falls into a different “family” of reconstructions, using an entirely different proxy network and methodology than TH2009.
Let’s start by trying to figure out the network from the sketchy information available in the PAGES PPT presentation, namely the following location map and legend which states that a total of 118 proxy series were used in the reconstruction (96 tree ring MXD series, 7 ice core O18 isotope series and 13 varve thickness series) with the locations shown below.
Figure 2. Tingley and Huybers PPT Proxy Location Map. The original caption says that proxy data was obtained from Konrad Hughen of Woods Hole.
MXD Data
From the pattern and count, the MXD version used here appears likely to be the gridded version of the Briffa-Schweingruber MXD data derived in Rutherford, Mann et al 2005 (also used in Mann et al 2008). For a long time, Briffa refused to disclose which sites were used in his various articles, but, as a result of prolonged quasi-litigation, this information became available in late 2008 in the wake of Mann et al 2008 and we have some dividends from this for TH2010.
The various MXD networks are described at a CRU webpage here. The locations of the 105 gridded series discussed in Rutherford Mann et al 2005 are here; there are precisely 96 series north of 45N and their locations match closely to the locations in the PPT location map as shown below. So for now, it’s a reasonable guess that TH2010 used the gridded MXD series of Rutherford Mann et al 2005 located north of 45N.
Figure 3. Emulation of TH PPT Location Map – see text for explanation.
Before we try to decode exactly how the fancy “new” methodology works, it’s always a useful precaution to show a simple average of the data for each class. A simple average of the 96 MXD series is shown below, showing the familiar “divergence problem”.
Figure 4. Average of 96 gridded MXD series north of 45N.
As a crosscheck on the above figure, the information webpage also identified 340 different MXD sites from which the 98 gridded series were derived. 330 of the 340 MXD sites have versions at ITRDB – a few series, mostly south of 45N are missing from ITRDB despite the CRU statement that all the series are at ITRDB. I also calculated a simple average of these 330 MXD series as archived at ITRDB yielding a similar looking graphic.
The MXD “divergence problem” has always been a problem in Team reconstructions and is once again in the Tingley-Huybers version. 96 MXD series out of a total of 116 proxy series in the Tingley-Huybers network go down, but the overall reconstruction goes up. Hmmmm.
Ice Core Isotopes
Kaufman et al 2009 recently reported on a network which included 7 ice core isotope records, shown in the above location map. 5 of 7 series seem to match TH locations, with TH apparently using a Mount Logan series (probably the old Holdsworth version) and a Penny Ice Cap, Baffin Island version, while not using two Greenland series used in Kaufman.
Kaufman refused to provide the supposedly “publicly available” data that he used, including certain annual ice core data that is not “publicly available”; my request for this data is currently under quasi-litigation at Sciencemag. In the meantime, the figure below shows the average of the seven Kaufman ice core series (decadal averages), which also go down. As discussed previously at CA, Fisher’s relatively recent Mount Logan ice core series (not included in the average shown below) also goes down in the 20th century.
Figure 5. Average of Decadal Kaufman Ice Core Isotope Records in SD Units. This is a direct average of ice core data as archived by Kaufman.
Varve Thickness
Varve thickness is something that we’ve discussed in the context of Kaufman et al 2009, which uses 9 series that one can count as varve thickness. TH2010 report the use of 13 varve thickness series (tho I can only locate 12 on their location map: perhaps a couple of sites overlay.) The Alaska sites seem to match Kaufman’s Iceberg Lake and Blue Lake; both studies have two sites in Baffin Island, with Donard in common, but TH perhaps having a different site in southern Baffin Island rather than Kaufman’s Big Round Lake; TH have a site in Svalbard, while Kaufman has a site in Iceland. Both have sites in Finland – I wonder whether TH use upside-down Tiljander where narrower varves are interpreted as evidence of warmth? TH have 6 or 7 sites in the Arctic Islands versus 2 in Kaufman. We’ve discussed problems with some of these studies already: e.g. inhomogeneity at Iceberg Lake and upside-down Tiljander.
Reviewing the Network
Tingley and Huybers develop a relatively complicated multivariate to extract a signal from proxy data – the classic Mannomatic situation.
The raw materials for the TH2010 reconstruction have the opposite problem from Yamal and Mann’s PC1 – they mostly suffer from the divergence problem. The 96 MXD series (out of 116) go down in the last half of the 20th century. The average of 7 ice core series also go down. The Tiljander series in its recommended orientation goes “down” (not due to climate). The Iceberg Lake series goes up but is plagued by inhomogeneity.
Quasi-Splicing?
Something else must be going on in the algorithm and it will take a while to sort through this new algorithm to see what makes it tick. Tingley has provided code for it, but hasn’t provided data. But before doing that, there’s one other aspect of the Tingley code that we need to consider. Tingley-Huybers also use 249 instrumental series. Tingley-Huybers (in their second methodological article) compare their method to RegEM. Maybe their method effectively splices an instrumental data blade with a nondescript proxy handle.
Otherwise, it’s hard to see how their method – Bayesian or otherwise – can get from the nondescript proxy network to a HS. I’ll collate and post up a network that is close to the Tingley network and maybe readers can analyse it with Tingley’s Matlab code.
Climate Audit 
121 Comments
I can’t help feeling that there is a certain amount of desperation and grasping at straws going on with the team now trying anything they can to pull hs’s out of thin air, keep up the good work Steve.
Is there any reason why all the data are from the Northern Hemisphere? And why the observations (as opposed to proxies) all appear to be land-based?
2nd para under heading “The TH 2010 Network” you refer to 118 proxy series- should this be 116?
So, am I right to say that if all these papers are using the same data then their results will be the same?
Regards
Mailman
Re: Mailman (#4),
They will be, if not identical twins, then closely related siblings. Having the same genetic code means that variation will be small.
As for the instrumental series: From the map it looks like it’s a gridded version, with all complications arising from that.
This is the divergence point between audit and peer review. The reviewer would submit this as a question for the authors to reply to. Then the edtitor would decide if the author reply was adequate. In audit, you track down the source of the discrepancy. Don’t tell me audit blogs do not perform a vital function.
Steve
“Maybe their method effectively splices an instrumental data blade with a nondescript proxy handle. ”
They wouldn’t do that would they? As DO has pointed out in his exchange with Jeff ID here.
“Hi Jeff,
Thanks for your comments. Some things we’ll naturally disagree on. The tone thing can be hard, for myself as well I realize — my thoughts on it, as well as things that I think get in the way of useful discussion and exploration, are here.
One thing I consider a priori unacceptable is derogatory comments about working scientists.
I was most interested in this part of your message:
accepting a tree as a thermometer
The fact is, no one thinks this. It is a straw man…..”
So according to DO dendros don’t think that tree rings are thermometers so why on earth would they even consider splicing them onto the start of an instrumental temperature record in the 20th century let alone actually do that? They just wouldn’t do that would they? And they certainly wouldn’t try to to include a ‘quasi-intrumental temeperature record reconstruction’ in IPCC AR4 would they?
David Appell must be feeling rather embarrassed and somewhat foolish at the moment – he ceratinly seems to be back tracking at the moment. I can’t wait for your next installment of ‘how to slay a hockey stick’ in 3 easy (lessons sorry I mean) CA threads. Perhaps they discovered Mann’s parallel mathematically universe in which it doesn’t matter if up is down or down is up (i.e. where the sign of number is irrelevant mathematically).
KevinUK
Re: KevinUK (#9),
David Appell is not backtracking so much as whining. Look at his 10/24 post at Quark Soup. It is pitiful.
======================
Re: KevinUK (#9),
DO refused to post my reply to this statement he made. Sent an email explaining that he partially agreed but didn’t want to deal with the ensuing moderation. I’ve saved a copy of my post and was considering putting it up so people could have the discussion.
Re: Steve McIntyre (#15),
Do you have a link to the code? I looked at the website where the submission is and have read it.
Re: jeff id (#19),
Delayed Oscillator was offended with my post in #19, he said I’ve mischaracterized his opinion and was a bad boy for mentioning his email at all.
Since I didn’t explain any of my own post on his blog, since he chose to snip my reply and and since I didn’t reveal any of his email, I wonder how it could be mis-characterizing anything. However, in the interests of playing nice, please ignore my #19 and assume it is completely factually inaccurate—- for some reason.
Re: jeff id (#21),
So assumed. Next.
============
Re: jeff id (#21),
And Tom P et al wonder why I would never comment there. Steve M, lucia, and Ben Hale do not use a heavy hand. Those that practice censorship can please leave the continent and go to Russia, China, Cuba, South Korea … and tell me from there how much you like it.
Re: bender (#23),
“South Korea”? Errr, hopefully you meant “North Korea.” Either that or I’m behind in world news. 🙂
Re: jeff id (#19),
There is a link to a package of matlab code halfway down on dr Tingley’s web page.
Re: KevinUK (#9),
I’ve certainly tried to persuade my bank manager that the sign of a number series is irrelevant but noooooo….
It’s like fighting a Hydra, a mythical being that regenerates its heads when they’re chopped off!
Re: Adam Gallon (#11),
One wonders if that is not a strategy by design. Flood the “frauditors” (great term, Hank Roberts!) with junk.sci so they are fully occupied in the month of November.
Having tracked Steve’s work for a few years now I can predict that the effect has nothing to do with Bayesian vs frequentist statistics and will arise strictly from an arbitrary “flavor choice” that will look a heck of a lot like confirmation bias. That’s the pattern. We’ll see if it holds.
It would be interesting to know if Mann was at the Pages 2009 Conference and if he made the comment that he has made in the past. Something to the effect, that no serious paleo would splice a temperature record to the end of a proxy reconstruction.
Re: AManuel (#14),
I don’t know for sure that the method is a sort of splice of instrumental records onto proxies. However, without doing something like that, as I said above, it’s hard to see how you can get from the proxy network to the stock. Please don’t go a bridge too far with what is for now a surmise about the effect of the method.
Steve, are you trying to trick us here? Have you plotted the Gridded MXD Average upside down? Not that there’s anything wrong with that. 🙂
Proxy construction and GCM construction have a lot in common. Select the trees/forcings that give you the best back-testing correlations, and run them forward. The proxies have had the full 20th century to generate divergence.* The models, being more recent, run off the rails within a couple of years of projecting forward looking results.
Until someone provides a believable explanation of how you get Fig 1. in light of Fig. 4-5, I will remain unconvinced.
*(Briffa’s splice of the dirty dozen is a direct reaction to this divergence-although in defense, he was only doing for the 20th Century what had been done in the proxy networks for the previous 1400 years, namely selecting the networks that provide the best correlations)
“It doesn’t matter why the science is bad”: to whom doesn’t it matter?
Thanks Thor, I saw the page but missed the link.
I am placing a bet that the gridded instrumental temperature got thrown in the hopper with the proxies to get the uptick, even though “no one would ever do that”.
Re: henry (#28), the increasing flood of papers is an attempt to sequester carbon and save us from ourselves.
It’s going to be interesting to see how they handled the MXD ‘divergence poblem’.
I’m amazed at how sophisticated we….er, you, are getting at this blog. We are already anticipating many of the significant issues with a study that has not even been released. But we’ve acquired a lot of experience over the past 4 years or so that may very well permit us to do so.
However, if these problems are that obvious, and they are; you have to ask why these issues aren’t openly dealt with and clearly refuted in each new publication. It’s like “we’ve moved on” but none of the battles have changed.
In fairness such a conclusion is in part dictated by the way this thread was set up. Now let’s see how well Steve M. and contributors to this thread have anticipated the areas of dispute in the actual article. Perhaps we should have a contest like we do in the Sea Ice threads. There could be numerous prizes.
Re: Barclay E. MacDonald (#31),
One of the main contentions of Steve has always been that the same “special” data is being used over and over. Do you think that if we use the same data over and over we should have to wait until a peer passes the thing before pointing out that it’s the same data.
Just how did Shweingruber’s MXD which has known divergence problems turn into a hockey stick? I may not be the sharpest bulb in the Hanukkah tree but something weird happened.
Re: jeff id (#34),
Barclay’s not being sarcastic. He’s genuinely amazed at Steve’s continual state of readiness. And he should be.
Re: bender (#36),
“Steve’s continual state of readiness”
This IMHO is what must p**s off The Team most of all. It must particularly annoy them that they are having to deal with someone (Steve) who has a much deeper knowledge of their field (multi-proxy temperature reconstructions) than themselves. I’ve always wondered what the Wizard of Oz thought about Toto. Now I know.
Judging from how quickly Steve turns around and refutes these continuous attempts at resurrecting these claims of ‘unprecented warming in the last thousand years’, Steve must be in a ‘continual state of readiness’. How else could he manage to do what he isn’t otherwise?
KevinUK
The (old) graph showing Yamal vs Polars is one of the most interesting I’ve seen here. To the extent your recent detailed critique of Yamal is correct that graph could be considered the “smoking gun” of tree ring analysis.
It’s very, very frustrating that the RealClimate folks won’t even try to mount a vigorous defense of Yamal as it’s clear how important it *continues to be* to so many studies and also clear from your work that it appears to have serious flaws.
I was always under the impression that Huybers, no matter his opinion about AGW, hewed to the statistical and climate proxy Straight-and-Narrow. He may have differences with Steve M., but they seemed to be honest differences. Now we get a proxy series from him, TH 2009, that has every single one of the no-no bugaboos that haunt proxy thermometry. We have strip-bark foxtail series. We have Yamal. We have Mann PC1, which is just foxtail redux. I’m suffering from cognitive dissonance trying to reconcile my S-a-N model of Huybers with the Huybers observable here.
So I offer my ad hoc fall-back save-the-theory rationalization. Huybers is propagating a rather cruel practical joke on the proxy climatology community. He’s out to show that approved-form reconstructions using even the most egregiously, blatantly malodorous ersatzproxies can still pass peer-review muster in a respected climate journal. That must be it. When he reveals all, there will be weeping and gnashing of teeth in the editorial ranks.
I took a look at the PAGES 2009 oral presentation. Slide 2 shows the instrumental surface record with a uniform uncertainty of (+/-)0.1 C. Ludicrous. The Bayesian model incredibly assumes a linear relationship between the proxy and the target climate observable. Nice for simplifying the calculations, a disaster for physical reality. The calculated confidence intervals can have no physical meaning.
Re: Pat Frank (#34),
Interesting theory.
oops #40 should have said “manage to do what he does otherwise?”
This review of an unreleased paper really reminds me of A Perfect Vacuum.
I notice from Tingley’s CV that he studied Maths and Physics at the University of Toronto. I guess we have something in common.
Re: Steve McIntyre (#38),
I give up. Too hard a question. Geoff.
Steve: So did I.
Re: Geoff Sherrington (#41),
Well, I think he was juxtaposing his figure with your #2&3 to reinforce the point that these proxies are all Northern Hemispheric. It’s worth thinking about but I suppose not terribly germane to this thread.
================================
Re: kim (#45),
There is a probability that some important global climate understandings will contiue to come from differences between the hemispheres. That was my point, made with attempted irony IIRC.
Re: Steve McIntyre (#38),
Hey Steve, please give the poor kid a break! He’s just finished his PhD and is exploring a rather interesting approach for reconstructing hemispheric temperatures. I was at Corvallis with the PAGES meeting when he presented these results and they seemed promising. Yes he has made some mistakes selecting the proxies and some of the assumptions of his technique are quite crude, but I think it is not useful at all for this young fellow to receive such a blow on his first research efforts.
Even Jean S. (#57 above) said “I would say this is the best Team approach so far.” I would add that he is not a “Team” member, but I kind of agree with Jean S that this is a promising aproach.
I think the tone on these recent posts on Tingley are examples why researchers in the field keep avoiding CA instead of participating more actively to exchange ideas, advice, etc.
Re: giano (#64),
Disagree. Who is his supervisor? And why is his supervisor not interested in doing good climatology? Maybe the kid’s not the problem. Maybe it’s a love of math and lack of care about climatological application. When working inter-discplinarily you must learn to care about the application. Sorry. Vehement disagreement is my gut response. We don’t need mathies embedded inside the echo chamber. We need the dendros to get OUT.
Re: giano (#63),
In all the posts that I’ve done here, I’ve very very seldom dealt with efforts by young scientists in anything other than encouraging terms (Ababneh, Kaufman’s MA students, Andrew Bunn, for example) and I sincerely regret that I didn’t take that into consideration in commenting on Tingley. My only excuse in this instance was that I got goaded by David Appell’s benediction of this new paper and didn’t consider Tingley’s age. In retrospect, I regret this. However, Huybers is a coauthor and he should know better.
I think that your point applies more to the previous post than to this one. I don’t see anything particularly offensive in this post – perhaps you can draw my attention to what offends you? Even in the earlier post, other than the link to Eric Clapton, is there anything in the head post that you find problematic? (Given Tingley’s age, I’d be willing to edit this post to mitigate this aspect of things.)
Or is merely pointing out that the majority of proxies go down something when the reconstruction goes up an observation that is excessively provocative in the context of a young scientist?
Re: Steve McIntyre (#66),
No problems, Steve. I was not “offended” by these posts. I just thought these posts were too focused on the things that Tingley did wrong without discussing with a bit more emphasis the potential applicability of his approach. I understand you can get quite angry when journalists publish and claim what they usually claim about each new “core Team” study, but is it Tingley’s fault?
I’m quite happy that you’ve been in contact with Huybers and am sure they will take and learn from many of the things you’ve pointed out here.
Re: giano (#85),
Where the BARSAT approach might apply would be to something like Antarctica as a way of avoiding the Chladni problem with Mannian PC decomposition. I find it hard to see any utility for it in proxy problems, because it, like RegEM disguises the statistical issues. Tingley has unfortunately gotten drawn into a Mannian way of thinking about proxies, but perhaps he can unlearn it. The other study is just another variant on CPS using familiar data and it’s hard to see where it goes. I’ll try to mention the Antarctic possibilities as a balance to what I regard as dim reconstruction prospects.
Re: giano (#63),
Peter Huybers has sent me a friendly email asking for comments, as the articles are not final. We had a nice dinner together in San Francisco at AGU in 2005.
Re: Steve McIntyre (#67),
A quick hint (should not take more than few hours to try), p. 10 (BARSAT I)
Suggestion: log-transform all tree ring data. As been noticed by several people here, tree rings (and chronologies) seem to follow rather a multiplicative than linear model.
Re: Jean S (#73),
And moreover: screening on the basis of MWP variance will unfairly bias against a warm MWP. To assume mean and variance are independent, such that there is no such screening bias, is a serious error.
Re: bender (#74),
What I would have liked to see from the Team is a composite plot of the proxies side by side with the temperature data for the 20th century. I thought the idea is to compare the proxies with the measured temperature data to see if the proxies were indeed good markers for temp so that we could reconstruct past temp data for the past when that data was not recorded. Now, the problem from what I have seen in many of the posts here is that the proxies and temps have a huge divergence problem in the later half of the 20th century. If this is so, how then can the proxies be used to accurately reconstruct
past climate behaviour? Now Tingley et al in one of his reports says that warm periods have both high means and variances, that both the
MWP and late 20th century have about the same variance, but the
later has a higher mean ( leaving out the lovely Polar Urals by the
way that was excluded due to too much variance). Unless someone can explain the divergence problem I don’t think the proxies can be used
with any precision to reconstruct the MWP climate. Now a possible
answer is that the tree rings act quasi-linearly up to the point of
optimum growth, which is a combination of high temps, adequate water,
plenty of sunlight, etc. If higher temperatures put stress on the
trees, their growth retreats and it looks like the climate is getting
cooler. It could be that this is what Polar Urals are telling the
Team, but they refuse to ask her out. It could be that the the best
indicator of warming climate is high variance.
This is the Southern Hemisphere from about 30 deg S to 90 deg S. Does it not matter?
http://i260.photobucket.com/albums/ii14/sherro_2008/CH30South.jpg?t=1256517149. Taken from Google Earth, with appreciation.
Steve: Not for this thread. As a general philosophy – let’s try to stick to the particulars of an individual topic.
Tingley’s methodological papers on “BARSAT” part 1 here constructs spatial covariance matrices on the basis of spatial decorrelation by distance – something that we discussed here in connection with Steig who, as we observed here earlier this year, ended up reifying significance to principal components that appear to be merely Chladni figures resulting from spatial covariance.
Unfortunately, it also appears that Tingley has adopted the Mannian method of jumbling temperature data with proxy data into one matrix.
Re: Steve McIntyre (#40),
On the surface the method seems fine to me. I rather like it in comparison to a less physically constrained RegEM approach. I’m not familiar enough with the math yet but it looks like an exponential decay surface fit of some kind.
I like this quote:
The irony of recent posts by Steve and RomanM and particulary the Figure 4 above, is that the divergence “problem” may be the rule, and may only seem strange because the few tree series showing an uptick in the 20th century such as Yamal and bcp are taken to be “correct” when they are not. Trees in far northern latitudes are often growing on thin soil over bedrock (because soil was removed by the ice ages) or thin soil over permafrost or have their roots in an organic mat (moss, usually). Any real warming will dry out these thin soils causing a decrease in growth. Over the long term, warming would cause deeper soils to develop and roots to go deeper after permafrost melts, but in the short-term the trees grow WORSE when it warms, in contadistinction to the dendro assumptions. These conditions can vary within a few hundreds of meters which can give local populations of responders, nonresponders, and negative responders, as has been shown by Wilmking for example. Simply picking the responders does not resolve the problem that many northern trees are not growing better as it warms.
Re: Craig Loehle (#44),
It could have just as easliy been a quote from Loehle et al. I’m not used to seeing the problem discussed that openly.
Re: Craig Loehle (#44),
One of my first comments at CA was on how divergence might be the norm on thin soils where temperature responses might be strongly nonlinear. I went one further and suggested that there could be a selection bias against the most drought-prone individuals. Which would be very relevant for trees trying to survive the MWP megadroughts. This means you would lose the evidence of nonlinearity, and the divergence would masquerade as a modern phenomenon.
Re: Craig Loehle (#44),
This is not exactly a new point here :). The “Divergence problem” has been an issue for a long time obviously.
This very nuance came up at the NAS presentations. Cuffey asked Mann about the Divergence Problem – Mann pointed to the Yamal chronology – I remember it distinctly and said: look there’s no divergence problem here.
I’ve obviously observed over and over the importance of the declining average ring widths in the Schweingruber network and that the odds of picking Yamal randomly were not very high.
Re: Steve McIntyre (#48),
Mann said that? Under oath? Bet he didn’t know there were only 10 trees during the divergence period.
Re: Steve McIntyre (#48), I know that YOU are aware of the divergence issue, I was just summarizing the irony for our readers.
what’s your point about the ice cores? It goes “down” you say… wait a tick, that means warming. So – the Arctic has been warming. Nice, salient point you make with that plot that you have arbitrarily going down…
Re: bent-out-of-shape (#49),
These are ice core “isotopes”. Steve didn’t label the y-axis, so I don’t know what the units are. Presumably a ratio or a difference. It matters which isotope is in the numerator and which is in the denominator.
…I guess his point wasn’t so salient b/c he didn’t label his axes… so which isotope is it? That would be really important – and may even be necessary, I would venture to guess. What do you think, bender?
Re: bent-out-of-shape (#51),
I think labeling axes is good.
Re: bender (#53),
I think so too. In fact, if some labels happen to pop up so that a reader can decide what the plot really means w/o an arbitrary orientation to the axes and some equally arbitrary text – then a reader may realize that the ice core data shows conditions getting warmer! Not just that they “go down”. …sounds kinda bad when you say it like that.
Re: bent-out-of-shape (#56),
hmmm, still no update yet on the non-labeled ice core figure that was misinterpreted and supports warming in the Arctic… I wonder if this will ever be acknowledged here????
Re: bent-out-of-shape (#60),
Here’s the irony, bent one. You want instantaneous updates here. You post on the same subject 4 times in a couple of hours. If you were posting at realclimate, asking them to, say, fix their broken graphs on the “Hey Ya!(mal)” thread (uggh, who writes their titles?), you wouldn’t get a single comment through because of their heavy-handed censorship. So why don’t you either calm down, or get your bent self over to RC and get them to fix THEIR graphs?
Re: bender (#61),
whatever, bender. You and your misdirection. I vehemently disagree with you. So much of the talk here demands immediate responses from scientists about all of the science that you “debunk” here. How hard is it to fix the axes and a some incorrect text. Instead there is a looooong post about a correction that still doesn’t do enough for you and many others. I guess a “shout-out” still isn’t enough… too funny!!
If you are *really* interested in the science, as you so humbly say, then you will answer my question about the isotopes in the ice core. Shrugging me off and telling me to go to RC is juvenile. And NOT scientific.
And if someone is too busy to scientifically address the comments about the isotopes in the ice core, the incorrect information should be deleted as to not mislead others who come here. That should only take 30sec.
Steve: the data is taken directly from the (first) Kaufman archive, which is in SD Units 980-1800. The information is “correct” to the extent that it represents what Kaufman reported.
Re: bent-out-of-shape (#65),
Answer your question? I didn’t see that you asked one. You made a point – four times now – and I preseume the blog owner will respond. Yes, it takes ~30 seconds to do it, and only 6 hours to find those 30 seconds.
.
How many seconds do you figure to delete the flawed hockey sticks at RC? 30?
Re: bent-out-of-shape (#65),
Ah, so that is not your plot. Now I understand why you can’t correct the labeling. But a more definitive figure caption and explanation would help. Given the SD of the variable, and the fact that ice is made of H2O, I would hazard a guess that d18O is the “isotope” that bender refers to. And yes bender, “neutrons” do exist – so no need for the quotations. Since you are not sure of what the plot means, perhaps we should talk about it a little more and not take the blog post for gospel.
The authors plot smaller values down and larger values of d18O up. No strangeness there. …but I think that more negative d18O results from warmer conditions, right? Isn’t that how the proxy records conditions? Therefore in that plot, warmer is “down”. So if the plot “goes down”, then it’s getting warmer. McIntyre, you are a paleo-proxy guy. I’m surprised that you missed that!
It’s not the direction of the change. You could reverse x & y and the plot would go sideways! More important is the scientific conclusion.
Steve: LEss negative O18 in ice cores is interpreted as warmer. Warm is up in the graphic shown.
“96 MXD series out of a total of 116 proxy series in the Tingley-Huybers network go down, but the overall reconstruction goes up. Hmmmm.”
i think it is a huge advance to be credited to steve mcintyre, that hockey-stick productions are now so well understood and so easy to refute. this gives good journals the possibility to debunk such stories before publication and embarrasing their own integrity.
I’ve now read a couple of times these BARSAT papers, and I do not see any major problems. Of course, there are always questions related to the used method itself (people do not like Bayesian methods, selection of priors etc.), and maybe more importantly, how good temperature proxies tree rings actually are and if their response can be even approximately described with linear models. The last point is actually mentioned in the BARSAT I paper. So unless something unacceptable (like truncating the MXD series; plus caveating that I haven’t gone through the code yet) turns out, I would say this is the best Team approach so far.
Steve, could you filter the “Gridded MXD average” with 9 point Hanning window (that seems to be used) and compare it to the filtered reconstruction [see figures 1 and 4 in this post (pre intrumental period)…]? It seems to me that after all this computationally expensive Bayesian modeling, all you get out of MXD data set is basicly the average of proxies…
Re: Jean S (#56), Jean if you get the code running it would be fun to remove a few bad men
Re: steven mosher (#77),
I think that you’re mixing up the Yamal-type study with the BARSAT study. The Yamal-type study is just a variant of CPS and I’ve got a long way through that one. The BARSAT study doesn’t use Yamal, but the divergence problem is suppressed somehow – probably by splicing the temperature record in RegEM style.
The BARSAT code indicates this. Actual gridcell temperatures go into the meatgrinder along with proxies. They assume that the proxies have a variety of properties that are inconsistent with divergence without showing that these assumptions are justified.
It does not matter why the science is bad and to whom doesn’t it matter?
Re: ejakulasi dini:#59 and dearieme: #20
It doesn’t matter for the purposes of this discussion or for this blog generally, or in a narrow sense, for science itself.
Obviously in the realm of public policy it matters and in a wider sense it matters for science as well, but those aren’t ordinarily the purpose or purview of this blog.
[Can we delete #16, as it is now orphaned?]
Re: Brian B (#62),
Thanks for clarifying. #16 was not intended to be a discussion point. It was a note to the moderator in response to a post that was deleted for impugning motive. The newbies don’t understand the blog rule against impugning motive. I pointed out that motive is not a focus of the blog, is not discussion-worthy here. But of course motive matters. Of course it matters what is motivating people to do bad science. It’s just not something that is discussed here.
Giano –
I agree that we should keep in mind that this is a young man who chose a research topic which he probably tried to complete to the best of his ability in order to advance or complete his academic training. He probably was not intending to become the the subject of blog battles along long-drawn lines, and should not automatically be recruited as spokesman for others of proven mendacity. Here’s the problem, though: if this just a young man adding incremently to scientific knowledge in humdrum fashion why is his effort immediately the subject of an article in Scientific American?
Re: Tom C (#68),
My guess is that he must have received a favorable reception to his method at the PAGES 2009 conference. Paleos (e.g. the PR Challenge) seem to think that the “problem” is getting a better multivariate method to “extract” the “signal”. Tingley probably got a good reaction because spatial autocorrelation (for example) is a more sensible assumption than temperature principal components reified as Mannian “climate fields”. Appell probably picked up on that.
99% of the response to our articles in academic community has been about esoterica like PC methodology, the number of retained PCs etc. Things that interest me mathematically, but which IMO are far less important for reconstructions than humdrum things like Yamal and Graybill strip bark bristlecones.
If you look at my comment at Andy Revkin‘s recently on reconstructions, my perspective is that the main “problem” is with the data:
Re: Tom C (#68),
The issue is not the guy. It’s the process in which he chooses to take part.
Re: bender (#71),
… a process where there is little concern for the quality and interpretibility of the source data.
… a process that seeks to downplay the level of uncertainty inherent in the science.
Showing science badness is not enough : one must tell that it is bad, explain how it is bad, then discuss why it matters.
I finally made heads and tails of your post and you did make the ice core plot. And you are correct, the ice folk say less negative d18O is warmer. So the plot you made of the ice-cores is 2000 years long and shows the cooling due to precession of the equinoxes up until ~1600, then pretty substantial warming from ~1900-2000. The same trend is apparent in Figure 3a of the Kaufman of the ice data.
Is the implication that there is no way to get a HS shape with ice cores not showing a strong warming in recent years??? not sure what your two short paragraphs are trying to say.
Re: bent-out-of-shape (#79),
THe ice core average ends up at a rather low level. It goes up in the first part of the 10th and down in the second half (according to the Kaufman data).
To the extent that the TH 2010 reconstruction – which is the topic of this post – is a sort of weighted average of properly oriented data, you can’t get a Stick from relatively evenly weighted MXD and ice core data, which account for 90% of the proxies. This seems self-evident to me from the plots and this is all that I’m saying so far. The main focus so far is trying to figure out the data.
Re: Steve McIntyre (#80),
Self-evident to you maybe, but not to me. Maybe you should be a little more explicit in the post. That’s probably part of the reason I couldn’t understand what point you were alluding to in those paragraphs. And what does the “first part of the 10th” mean???
Re: Patrick M. (#81),
that’s what I was doing. Mulling over the ice core data. Did you read #79?
Re: bender (#82),
I’m confused – why do I have to apologize? It was/is an unclear post about the ice core data. Also, I have no affiliation/involvement with RC: why would you expect me to fix their graphs? Pretty random question to ask someone posting on this blog.
Steve: no need to spar. I agree that the graph was not clear and I fixed it. OK?
Re: bent-out-of-shape (#83),
snip – (Steve: I try to be evenhanded in snipping potential foodfights. As you point out, they are unattractive editorially.)
But it would be worth noting what giano wrote. It is nice that Huybers e-mailed and asked you for your comments. That guy got a genius of the year award. Pretty outstanding. He got it for *really* knowing things. …not saying he does on a blog somewhere, like those who post here quite often. You have shown that you know the material and statistics. It’s time for you to ask yourself if it’s worth feeding your readers everyday with tons posts or making a lasting contribution. …and followers like bender, tho they are loyal, are not helping you accomplish what it is you really want to do.
Re: bent-out-of-shape (#87),
Huybers may truly be brilliant, but I’m not much impressed by awards. Remember Michael Mann won an award or two a few years back and Al Gore and Barack Obama have won Nobel Prizes. Science does not bow down to authorities… or to award winners.
Re: bent-out-of-shape (#79),
Maybe Steve is trying to look at the data as a scientist would, carefully and without bias? Perhaps you are new here? Steve will often use blog posts as an open diary of what he’s thinking about a topic, without necessarily having a conclusion in hand. The more capable posters will often join in on the “brainstorming”. Of course some posters with agendas, (either way), will try to push conclusions, (either way), but they usually get brushed off.
snip – point made, but let’s stop the incipient foodfight right here.
Steve: c’mon guys. no fighting. I’m offline overnight.
snip – thanks for the support, but lets’ nip this potential food fight in the bud.
I’m rather looking forward to this discussion of Tingley-Huybers. Huybers is without a doubt brilliant (his Pleistocene palaeoclimate work is first class and his MacArthur Genius Award well deserved), and he seems happy to engage. This could be very interesting indeed.
I saw what was written and did not reply. I am going to entertain giano’s hypothesis that in my absence there will be a flood of dendros to this blog to regale us with their luminous insights. I thought I had been filling a void for 4+ years. But perhaps giano – even though he obviously hasn’t read the blog to witness my past contributions – is right. With Hu, RomanM, and now some other statisticians adding to the dialogue nowadays, I really don’t have that much to offer. So have it it, dendros. bender is out of your way. If giano is wrong, he can apologize. We’ll run the experiment until Christmas.
Re: bender (#91),
You know bender I never postulated such “hypothesis”. I just pointed out that your arrogant way of responding to almost every single comment made by outlanders was really tiring and making more damage than good to CA. I’m sure the exchange of ideas and comments will be much more productive now that people (dendros and non-dendros) will be able to post without an almost immediate, almost always non-constructive reply from your side.
Re: giano (#92),
Why don’t you join in on discussions for a while before picking fights. Prove your worth. Bender’s contributions to this forum dwarf yours. He’s one of the reasons I mostly lurk here. Your contribution to this point seems to be a whiny admonishment of Steve and a personal attack on bender.
Re: bender (#91), Bender, Bender. Come back, Bender!
Bender, in my experience, only responds harshly to three different sorts of people:
1.Those who show their work but won’t let anyone else see the data or methods used to produce it.
2.Those who show their work but it’s nonsense.
3.Those who push an agenda on either side of an issue inappropriately, especially an OT one.
Does he go a might overboard toward any of the above at times?
Judging by the occasional snip he gets I’d have to say yes.
Does any competent, honest dendro or other scientist have anything to fear from bender?
Judging by the respectful hearing he’s given the real scientists of any persuasion who have posted here I’d have to say no.
Bender is something like a cat; usually he is quiet and fairly docile, but occasionally he makes a bit of a mess digging the rats out of the pantry.
well bender if you’re taking a break I will miss you. I’d like to say thanks for your holistic knowledgeable and passionate stance. And compassionate and funny. And even if tending to OTT, you always are more than willing to say sorry when needed. You see through people. You see through the science issues. You see through the trees from the forester’s POV. You cut to the real issues. Like Steve, you’re on the ball. I don’t know how you do it. Ah, yes, you’re a cat, I forgot. Nine lives.
I’m concerned with the NASA GISS temperature message on slide 2 of the BARSAT presentation.
The map shows anomalies around Yamal of over 2.5ºC above the global “mean” value which looks like between 1940 and 1980 or so (doesn’t define). But I have 20 GISS Yamal-PU area records in front of me and the average at Yamal appears to scarcely rise 0.5ºC over Yamal’s 1940-1980 mean. Perhaps 2008 was up in Yamal by 1.5ºC. So what?
The GISS global temperature increase over the last century is only ~ 0.7ºC. If the global mean now looks higher than the 1930’s, unlike what the Russian records show, has it been distorted by UHI?
Or have I missed a trick?
Has this GISS record been stirred into the Bayesian soup?
Bender When you take a break you always come back gently and get gradually more frustrated with the likes of TOM PiPi and giano. So my friend take the break and I look forward to your comments after Copenhagen.
If Bender did not exist, Steve would have to invent it.
Giano, Bender regularly takes a break – he comes and goes in waves, and he has been a presence here from the early MBH99 days. When he is in hiatus all that happens is that more people get away with a lot more nonsense. I’m surprised that you cannot see the underlying fairness of his comments, and that you appear to have failed to appreciate the depth of his knowledge and his many substantial contributions on matters of fact. From what I can gather, bender’s other life is senior academic in a related field – he reminds me of the profs I had in the 1970s/1980s who were trained in the 1950s – intellectually ruthless in the pursuit of truth, and merciless in tearing apart sloppy thinking. We need more of this, not less.
But bendie why? Now I have a saddie. 😦 I guess you really are a fan of the extremes and live by it. respect.
Re: bent-out-of-shape (#102),
Silly sarcasm
Your post #87 gave it away – you prefer all the analyses, hard doubts, debunking and critical questioning to be kept out of the public arena
Ho hum
Re: bent-out-of-shape (#102),
It’s mid-term crunch-time, and giano’s hypothesis – that there will be a flood of dendros in the wake of my leaving – offers serendipitous timing for a much-needed break. It is too bad the tail end of my activity coincides with your entry to the blog. But say something enticing and perhaps I will not be able to resist.
.
Yes, I am a cat who hates rats.
mb – I did. Check the comment about the Indonesian sediment cores.
Re: bent-out-of-shape (#105),
Enticing, but not enticing enough. Maybe “provocative” is the better word.
(What bad timing, to announce my intended departure just when Briffa makes a move. Ah well, the stack ever grows.)
Re: minimalist bender (#106),
OK bender, I’ll try to end this silly “fight” in the hope that you come back as a full bender instead of the mere minimalist that’s been visiting CA these days. Please accept my apologies for going a bit too far with my comments regarding your role here at CA. It was not my best day and should have probably waited a bit longer before posting those comments the other night. My main point remains though that a respectful exchange of comments and ideas is essential if we want CA to continue to grow and involve more people from the research fields that are the topic of daily discussions in this blog. And please also accept that the “no bender = flood of new dendro commenters” is your invention not mine! :))
Re: giano (#107), Ha! I see bender is already back and with full power on the new Briffa-Yamal post.
Re: giano (#108),
Hi giano. I won’t be battling any dendros. And it’s just for three days. I had no idea Briffa was going to reply this morning. I’m totally not interested in scrapping.
Re: giano (#107),
Noted and appreciated. 3 days, no scrapping.
Re: minimalist bender (#106),
yeaahhhh…. I thought as much. Unfortunately, too much attention is paid to these hemisphere reconstructions – on both sides. They are about flashy press coverage – for both sides.
What about the mechanisms that are causing these centennial-scale changes? That is what should be audited. The name of this site should be “StatisticsAudit” or “DataArchiveAudit”. And it’s not that those two items are unimportant, or are not a worthy cause to fight for. They are.
But auditing the climate issues this site is not. And not b/c you don’t feel like posting. It’s b/c no one else did.
Re: bent-out-of-shape (#111),
Hi bent. This is Steve’s blog. The topics covered are the topics that interest him: primarily multiproxy recosntructions and peer review. If you want other aspects of climate science audited, go for it. Start your own blog. If someone were to start auditing the climate modeling literature, I’d lurk for a year and gradually start to comment. If you could have your own climate audit, what would you cover?
.
Have you viaited lucia’s Blackboard?
Re: bender (#112),
The name should reflect what is being provided… just saying.
No, haven’t seen Lucia’s. I will check it out.
I would cover different paleoclimate reconstructions. What are different areas experiencing? Does that help us know what is driving the changes? I also find it especially annoying that most people concentrate on the US/N. America and Eurasia. I understand that this is where the most records are – b/c it’s easiest to go in your backyard and collect some records. …but there’s a lot more to this planet than those places.
Re: Curt (#113),
Yes, I agree completely. But how useful are the N. hemisphere reconstructions/averages? Why is everyone so obsessed with them – either for or against? …maybe it’s b/c the general public can easily understand them. Compare two temperatures – MWP vs. CWP – and arrive at an answer. It seems like they are geared toward a public with a 30 sec attention span. Sound bites.
How about which areas were affected during these climate changes? And how? Like the S. hemisphere? Or Africa? Or the tropics? Or India and the monsoon? Warmer or colder? Wetter or drier? That’s why I posted under the Indonesian sediment cores that bender commented on. …and no one even blinked an eye at it.
Re: bent-out-of-shape (#114),
The multiproxy blogosphere is a little bit pre-occupied right now with a pretty significant storyline: Briffa’s Yamal. That’s not hard to understand. The NH is land-dominated, hence the richness of records there. That’s not hard to understand. The treeline in the NH is extensive. Glaciers are extensive. The Arctic is experiencing unprecedented warmth and sea ice melt, we are told. Kaufman et al (2009)focuses on the Arctic.
.
What would you like to discuss about the SH that would out-rank these other topics? Of course the SH is important. But what about it is topical?
Re: bender (#115),
“Of course the SH is important. But what about it is topical?”
Bushfires that have burnt entire small towns to death (whole families melted to death while fleeing in their cars), coastal erosion that is a few metres from destroying $billions of property
The “meeja” here is having a field day blaming these ongoing events on our greedy consumption of fossil fuels. That such events have occurred previously in quite recent geological time is a very topical point, in my view, and this needs to be counter-driven
Re: ianl8888 (#117),
weather
done
next
Re: bender (#115),
you pointed out a good article about ocean sediment cores from Indonesia. I responded. No one else did. Not even you after calling it “provocative”. So I’m like WTF. Not hard to understand.
BTW, I may have missed it: did anyone talk about precession changes during the last 2000 years when discussing the Kaufman 2009 paper – or just about proxies and statistics. If so, I’d be happy if you could point that out, b/c I couldn’t find it. …just wanted to do some actual climate auditing.
Re: bent-out-of-shape (#119),
I don’t think that was discussed per se. Steve and I noticed that the time frame of the simulation was off. But, as I said, we stopped auditing that paper when everybody got yanked by Briffa and Yamal. Go ahead and give us a read of the paper. Say something provocative about it. Intrigue us.
Re: bender (#120),
Intrigue you? …wow. You got moxy, massengill-bolsa.
How about the statement in this post that the ice core record “goes down”. Of course it gets cooler. Due to precession of the equinoxes, JJA (June-July-August) insolation is decreasing over the last 2000 years. This is quite evident in the ice core data as McIntyre points out by “it go down”.
What isn’t addressed in this post is the uptick during the last 100-150 years or so. All things being equal, it should keep on “going down” as JJA insolation continues to decrease – just as it did for the last 2000 years (Figure 3F). Both the ice core and lake sediments show it. We’ll leave out the trees to make you happy. What is being argued about here is the length of the “blade of the hockey stick”. Why is there even an inflection at all?
What’s the cause of the increase in temperatures over the last 150 years in both lake and ice archives??? Feel free to throw out any bad datasets and reaverage.
Re: bent-out-of-shape (#119),
There’s an art to baiting auditors. You’ve got to find that hook. Your challenge is you have tons of competition right now. Lots of worms out there.
Re: bent-out-of-shape (#111), Until we understand the magnitude and character of these centennial-scale changes, how can we make much progress on the mechanisms? If the “traditional” assessment of these changes is correct (e.g. Lamb et al, with strong MWP and LIA periods), then we must posit (pre-industrial) mechanisms (forcings, internal variations) significant enough to explain them. If the “modern” (post-modern?, i.e. “hockey-stick”) assessment is correct, then we come to a very different set of mechanisms. First steps first.
Personally, I am less interested in whether the MWP was warmer than the CWP or not than I am in the magnitude of the pre-industrial changes. From what I can see, the dominant climate models cannot explain the centennial-scale changes of the traditional assessment. It doesn’t even seem to me that they can even explain the decadal-scale changes of the 20th-century temperature record.
When Steig et al was published the SH was highly topical.
3 Trackbacks
[…] Climate Audit points out that Tingley and Huybers also have an unpublished 600 year reconstruction that […]
[…] latest attempt to resurrect the hockey stick by Tingley and Huybers was quickly dismissed as more of the same using some statistical manipulation. Continued denial of […]
[…] a special feature appearance, Mann’s PC1 (though MBH98-99 are not cited). [Oct 24 – Also see followup post here; it appears that Appell was discussing another unpublished Tingley and Huybers not […]