Arctic lake sediment series have been an important component of recent multiproxy studies. These series have been discussed on many occasions at Climate Audit (tag), mostly very critical. PAGES 2017 (and related Werner et al 2017) made some interesting changes to the Arctic lake sediment inventory of PAGES 2013, which I’ll discuss today.
Some prior Climate Audit criticisms have resulted in withdrawal or major changes or lingering controversy, including:
- the controversy over Mann’s use of Korttajarvi (Finland) varve thickness and XRD proxies (tag) due to two factors: recent disturbance of varves by agricultural runoff and road construction; use of the data upside-down to the interpretation of the author. While Mann denied any error and refused to correct, Kaufman et al 2009, who had followed Mann’s incorrect orientation, grudgingly conceded the error after it had been pointed out at Climate Audit and issued a corrigendum inverting the orientation of the series (also reversing the orientation of four other series (of 23) in the corrigendum;
- the orientation of the Hvitarvatn varve thickness series in PAGES 2013 was immediately criticized at Climate Audit in April 2013 (here). Subsequently, McKay and Kaufman (2014) conceded the error and issued an amended version of their Arctic reconstruction, but,
like Mann, refused to issue a corrigendum to the original article.[Correction July 29, 2017: on October 7, 2014, immediately after publication of McKay and Kaufman 2014, I wrote to Nature, pointing out that McKay and Kaufman 2014 primarily addressed errors in PAGES 2013 (not new information) and urged that they issue a proper Corrigendum to PAGES 2013. In November 2015, over a year later, PAGES 2013 issued a belated Corrigendum. Neither Nature nor the authors notified me of this and I was unaware of this until a comment in this thread.] - almost immediately after publication of PAGES 2013, I pointed out that their Igaliku series (which had a huge hockey stick) was contaminated in its modern portion by agricultural runoff and erosion. I recall that Nick Stokes vigorously denied that this constituted an error. McKay and Kaufman 2014 partially responded to the problem by truncating a few data points towards the end (but not fully extinguishing the problem.)
- I observed that the Kepler Lake (Alaska) d18O series was structurally similar to the Mount Logan (Alaska) ice core d18O series, about which I had criticized its ex post exclusion by PAGES 2013 on spurious grounds – the series went the “wrong” way and was excluded due to “regional” effects. Instead of re-instating Mount Logan (according to principled ex ante criteria), PAGES (McKay and Kaufman) instead excluded Kepler Lake.
Other prior CA criticisms of Arctic sediment proxies included:
- inhomogeneity of the Iceberg Lake varve thickness series (tag) due to changing physical configuration of the moraine-bounded lake;
- arbitrary exclusion of the early (pre-730) portion of the Blue Lake varve thickness series (tag), where the alleged inhomogeneity was simply that the climate was believed to be “warm with precipitation inferred to be higher than during the twentieth century”, not a physical disturbance.
Some of these issues had material impact on underlying reconstructions e.g. the notorious Mann et al 2008 “no-dendro” series or Kaufman et al 2009 (here).
McKay and Kaufman 2014 (PAGES)
McKay and Kaufman 2014 discontinued the following three series used in PAGES 2013 on the grounds that there was “insufficient evidence that they were sensitive to temperature”
- the Kepler Lake (Alaska) d18O series mentioned above;
- a particle-size series from East Lake, Southampton Island in the Canadian Arctic Archipelago;
- a diatom (Fragilariopsis cylindrus) series from Holsteinborg Dyb in west Greenland.
They grudgingly inverted the Hvitarvatn series.
Sections of the following two sediment records were truncated on the grounds that the record was not “sensitive to temperature” during these intervals:
- Blue Lake varve thickness prior to AD730;
- Igaliku pollen accumulation rate after AD1920
A 50-year dating error in the Lone Spruce (Alaska) BSi series was fixed.
PAGES 2017
Four more PAGES 2013 Arctic sediment series were rejected in PAGES 2017, while two series used prior to PAGES 2013 (but not in PAGES 2013) were re-instated.
The following four PAGES 2014 Arctic sediment series were rejected in PAGES 2017:
- the Iceberg Lake (Alaska) varve series (previously criticized at CA tag) because of its “unclear relation to temperature”
- the truncated Korttajarvi X-ray density (XRD, a form of grey scale) see CA tag, originally used in Mann et al 2008 and Kaufman et al 2009, now also said to have an “unclear relation to temperature”
- the previously truncated Igaliku series (CA tag)now removed in its entirety due to its “unclear relation to temperature”
- the Lone Spruce (Alaska) BSi series (introduced in PAGES 2013, but only mentioned in passing in CA posts) was also removed due to its “unclear relation to temperature”.
The SI to PAGES 2017 also contained the following lengthy quality-control comment on Hvitarvatn, which had been a source of PAGES 2013 embarrassment:
Based on studies of glacier mass balance and glaciology in Iceland (e.g. Bjornsson; Flowers), Icelandic glacier fluctuations are dominantly controlled by changes in melt season temperature. Glacier fluctuations influence the production and transport of eroded material and the eventual deposition of this sediment in a downstream basin (i.e. a proglacial lake). .. On short timescales (seasonal, annual, inter-annual), changes in sediment accumulation can be driven by many factors and we can all agree that identifying individual controls is messy. But on longer timescales (for example, centennial timescales, … I would argue strongly that changes in sediment accumulation are driven by changes in glacier size. This is laid out in Larsen et al., 2011 QSR. We subsequently expanded on this initial study to: 1) include the whole Holocene (Larsen et al., 2012 QSR attached, which demonstrates a clear “8.2ka-event” signal and subsequent Neoglacial onset), and 2) by measuring varve thickness in multiple cores along a lake transect and tying the core data to seismic stratigraphy (Larsen et al. 2013 EPSL attached). This latter work demonstrates that the trends in sediment accumulation are consistent and observed throughout the lake basin. Given the available data, I feel comfortable summarizing as follows: Icelandic glacier fluctuations are dominantly controlled by summer temperature. On longer timescales, fluctuations of the Langjokull ice cap can be reconstructed from changes in mean varve thickness at glacial lake Hvitarvatn.
Previous comment: QC failed: article states “sediment flux to Hvítárvatn is dominantly controlled by the integrated rate of sediment production by erosion beneath Langjökull, modulated on annual to decadal timescales by the efficiency of the subglacial fluvial sediment delivery system.”, variability function of proximity, absolute values function of sediment availability. This is _not_ temperature!; QQ by PF not passed
The two paragraphs seem inconsistent. They did not discuss or attempt to explain the similarity of Hvitarvatn and Big Round varve thickness series.
The following two series were re-instated:
- Soper Lake, a varve thickness series used in Mann et al 2008. It was too short for either Kaufman et al 2009 (1000 years required) or PAGES 2013 (minimum AD1500 start). However, it has a pronounced Hockey Stick shape over the shorter period. I think that the policy of Kaufman et al (restricting to long series) is by far the better way to achieve interpretable results and regard this re-instatement as quality deterioration.
- Hallett Lake, which like Lone Spruce, was an Alaska BSi series (one with slightly lower resolution). It had been used in Kaufman et al 2009 but dropped in PAGES 2013. It’sfar from obvious why one Alaskan BSi series is now held to be temperature sensitive and not the other – especially when the opposite was concluded in PAGES 2013.
The Hallett Lake BSi series extends back to the mid-Holocene and is shown below to keep changes in the last millennium in perspective. The Hallett BSi series does indeed have somewhat of a Hockey-Stick shape (which might have explained why it was preferred to Lone Spruce BSi), but on a Holocene scale, the blade of the stick is inconsequential – a point that can be missed when multiproxy techniques first convert series to SD units.
PAGES 2017 introduced two new low-resolution alkenone series, both by D’Andrea:
- Lake E, a Greenland Lake adjacent to the Braya So alkenone series already in PAGES2013. (The original author even combined the two series in a version).
- Kongressvatnet in Svalbard (16.1 years)
Werner et al (CPD 2017)
Werner et al (CPD 2017) is an Arctic reconstruction by PAGES2017 authors. Strangely, it rejects three more Arctic sediment series, each of which had been involved in previous controversy:
- it rejected Hvitarvatn (a non-HS series) on the grounds that its “annual and centennial signal inconsistent”. It didn’t say how it arrived at this conclusion.
- it rejected Blue Lake (which had very elevated first millennium values) on the grounds that it had a “very nonlinear response, short overlap with instrumental, unclear interpretation”.
- it rejected Lehmilampi on the grounds that “exact interpretation unclear from original article”. Yet it retained Nautajarvi from the same authors, even though its “darksum” series is, if anything, harder to interpret.
In their reconstruction, they elected not to use 9 series on the grounds that they lacked annual resolution: six chironomid/midge reconstructions (Hamptrask, Lake 4, Pieni-Kauro, Hudson, Moose, Screaming Lynx), three alkenone series (Braya So, Kongressvatnet, Lake E) and the one remaining BSi series (Hallett Lake).
The net result is that the sediment series in Werner et al 2009 reverted back to five series from Kaufman et al 2009 (Donard, Big Round from Baffin Island; C2 and Lower Murray from Ellesmere Island; truncated Nautajarvi, Finland) plus the short Soper Lake (Ellesmere Island) series to help with the HS.
Taking Inventory
The inventory flows of Arctic sediment proxies are summarized below. 32 “different” (non-isomorphic) series were introduced in the four studies as “temperature sensitive”. (For the purposes of this inventory, flipped versions are treated as one series.) 16 of the 32 series were rejected in a subsequent study as not being “temperature sensitive” after all. This is a very high casualty rate given original assurances on the supposed carefulness of the original study. The casualty rate tended to be particularly high for series which had a high medieval or early portion (e.g. Haukadalsvatn, Blue Lake).
Not only is the number of surviving series (16) a discouraging proportion of the opening inventory, but even the opening inventory was itself culled from a much larger population of lake sediment series. Making matters worse, because the inventory of proxies changes only slightly from study to study, the networks of sediments in each study are not independent, as opposed to slight variations. Because data snooping and ex post selection are endemic, little to no credence can be given to the (very slight) HS displayed by any composite. (This is not to say that the “true” answer is a non-HS, only that the answer is tainted.)
In trying to get a more informed understanding of Arctic proxies, I’ve found it helpful to examine Arctic lake sediments over a Holocene perspective and to examine multiple series (e.g. varve thickness, magnetic susceptibility, BSi) from a single site at the same time. I think that it is most helpful to work with proxies which are available at multiple sites (PAGES too often uses singletons). I also think that it is best to work out from the best understood sites – a principle used in the mineral exploration industry. I have some notes on some sites that have particularly interested me (Hvitarvatn, Big Round in particular) and will try to write them up some time.
Climate Audit 
85 Comments
Excellent and rather devastating analysis Steve!
Thank you for the detail.
Steve:
It seems to me that your conclusion demands a response from the authors:
“Because data snooping and ex post selection are endemic, little to no credence can be given to the (very slight) HS displayed by any composite.”
Is this not an archetypal example of data mining?
So we can now make explicit the cooking of data sets?
it’s a gradual and slow-motion data mining. “Cooking” also implies a malicious intent that is not applicable to JEG anyway and not helpful to discussion.
In my opinion, there is a definite bias, but it’s mostly unconscious or rationalized with good intentions. I wouldn’t necessarily say the same thing about someone who willfully withheld results so as not to give “fodder to skeptics”, but that’s not JEG.
Steve:
Willful ignorance regarding proper data collection and treatment, at this point, six decades after the prevalence of specious and spurious correlations
was first fully appreciated by the econometrics community is unacceptable.
We still have to deal with people who haven’t absorbed the first semester’s worth of introductory statistics?
IMO, this behavior isnt unintentional.
At some point the doctrine of constructive knowledge must come into play.
If one is going to use the tools of statistical analysis, it is incumbent upon that person to have at least a rudimentary understanding of the tools one seeks to use, no?
It is simply impossible to practice such malfeasance unintentionally now.
For example, Bonferroni corrections have been employed for almost 60 years now to correct (at least partially) for this type of data treatment.
Scalia wrote that first principles are important. The is true when applying statistical analysis to data.
Violating these principles renders whatever results one may get worthless.
Ex posts selection of data, the use of autocorrelated time series without acknowledging their characteristics?
This is “kosher”?
Using PCA on a bunch of components all of which are hypothesized to be independent of each other, yet simultaneously are all also valid proxies for the same phenomena?
Steve, can principal components be orthogonal to each other, and all still be proxies for the same thing?
Can there be a believable hypothesis for just why one set of proxy data would capture some of the variance in temperature, while another supposedly valid proxy would capture some other aspect of climatic variance?
The whole charade that is PCA “climate science style” isnt just one data snooping exercise?
Really?
These behaviors are widespread and pervasive in the dendroclimatological community.
These guys are not only ignorant of the issues presented by their malfeasance, they are prideful about their ignorance.
Throwing out data that doesn’t conform to one’s hypothesis, cannot be ascribed to “good intentions” but thats my opinion.
In any other field of scholarly endeavor, this type of behavior would warrant sanctions, not be rewarded.
When people with graduate degrees do things that would warrant a failing grade on a data collection and treatment protocol assignment given to a first semester statistics student, and then crow about it, I think that your assumption of good faith is charitable.
Youre a better man than I am MacDuff.
“Steve, can principal components be orthogonal to each other, and all still be proxies for the same thing?”
That was one of my first questions long ago.
However, if proxies are spatially autocorrelated (as some of them are), the principal component decomposition will yield eigenvectors with pretty Chladni patterns. There are some old (and IMO very good ) CA posts on this.
Thanks for taking the time, I have some recollection of those posts.
I just went and reread your post
That some would choose to denigrate you only confirms the supposition that they haven’t read too much of your analyses.
I can understand how the Eigen vectors calculated over some interval might appear to be orthogonal, but the underlying phenomenon cannot be independent of each other, if the proposed physical relationship between the various proxies and climatic variance exists.
IMO, this implies that one is examining a statistical artifact, an evanescent pattern, not some fundamental physical reality.
What is the rationale behind the hypothesis that some types of systemic climatic variance is picked up by particular proxies and other components of the same systemic variance are picked up by different proxies?
Thats is,
is there evidence other than that gathered by running every possible permutation one can think of until one gets an ensemble of components that looks like the phenomena one wants to model?
if a “proxy” is a proxy , it is y= a*x +e. If one has a network, the x will be best estimated by the PC1, but almost any all-positive weighting will recover something close to the “signal” x. But if one has a clump of (say) bristlecone pine chronologies which are close to each other and a distinctive pattern and insert into a matrix of signal plus noise (by construction), they will form a distinct pattern orthogonal to the signal. I did some experiments with bristlecones and Chladni that were interesting. I forget whether I wrote them up.
The amusing property of Mannian principal components was that it would seize on such HS shapes in preference to a real signal. If the “true” signal was a HS shape, it would exaggerate the shape. Its use didn’t mean that the climate didn’t have a HS shape, only that it was an erroneous way of seeking it.
Thanks again Steve.
Im sure that one can find proxies (or a network of proxies) that are orthogonal to temperature variance, and to me it makes sense to combine groups of individual time series if they are strongly correlated with each other…..and if there exists some physical explanation for combining them..(same species of plant, same type of dendroclimatologial proxy….same type of paleoclimatological proxy….
I guess I didnt phrase the question write precisely…
How can all of the principal components be orthogonal (independent) to each other if theyre all proxies for the same phenomenon (variance in climate)?
David Eisenstadt :
PCs are orthogonal by construction, but the first several PCs can represent several different, cross-correlated climate variables. Even if the climate variables were independent, there would not necessarily be a one-to-one correspondence between PCs and variables — Temperature, for instance, might be one linear combination of the first 3 PCs, while precipitation was a different linear combination of the same 3 PCs. Significant PCs may also to some extent reflect non-climate variables like type of proxy (d18O vs Mg/CA, stripbarkedness, Tiljanderness, etc). The advantage of PCs is just that they reduce a vast data set down to a handful of summary statistics that presumably contain the information above noise, without “peeking” at correlations with temperature.
One objection Steve has had to PCs is that they will in general place mixed signs on the underlying series, so that some proxies may get a positive coefficient in a a reconstruction while others may have negative coefficients, even if the physics of the proxy is that the coefficient can only be positive. I have argued that this can be corrected by dividing PC1, say, into two components, PC1+ and PC1-, depending on the signs on the proxies, and treating these like separate variables.
PCs of course do not correct for prior screening or cherry-picking of the proxies — GIGO.
Dear Hugh…
Thanks for taking the time to try to educate me…
What is being created is a reconstruction of temperature, no?
Do for example, the people involve in pages 2017 claim to attempt to explain precipitation?
So, PC1 explains more of the variance in temperature than does PC2, or PC3…but the phenomenon they all explain is variance in temperature… no?
So, variance in temperature may be some linear combination of principal components, all of which individually are held to be valid proxies for variance in temperature…
What is the physical basis for the weightings that the various proxies are given?
There insnt one, there cant be one.
Its curve fitting.
Arent all of the PCs used to explain variance in one dependent variable, temperature?
What am I missing?
David Eisenstadt:
In the case of tree rings, we would expect them to respond to moisture, aerosols, insolation, CO2, and maybe temperature. It’s a purely empirical issue what the relative strengths of these effect are, which makes it an interesting statistical issue. PCs are one way of reducing the data to a manageable number of series that hopefully pick up any signals that may be present without necessarily cherry picking. Admittedly curve fitting, but that’s how a lot of empirical science works.
In the case of glacial d18O ratios, there is a physical relationship that can be measured in the lab, again empirically, but with high precision. However, it remains another empirical issue just how well this relationship shows up in the field.
As for lake sediments, varve thickness is sometimes positively and sometimes negatively correlated with temperature, depending on how the lake is fed. Furthermore, the instrumental period is often contaminated with modern development and agriculture, so that it strikes me as a highly suspect proxy. But maybe there’s something there. There’s also biogenic silicates (BSi), but I don’t know how they are supposed to work.
Hugh:
thanks again.
I have some familiarity with the underlying assumption regarding proxies…
I suppose Im conflating a few distinct issues:
1) Mannian PCA, and the independence of the varoius PCs used, and
2) the collection and treatment of data by the paleo community, and
the ultimate utility of those proxies.
I have some experience with needing to utilize proxies; I spent some time in the 1980s developing multivariate macroeconometric models….almost all macro data is in fact some sort of proxy, and most econometric data is noisy, autocorrelated, and often heteroskedastic.
For me, the list of confounding and unquantifiable factors other than temperature that affect tree ring width alone render them unreliable at least for the purposes they are now used for….
That inspires another question, one that’s probably been asked before…
If the temperature profile (whether regional or global) is matched by a tree proxy, it is said then that tree is sensitive to temperature and a valid proxy, backed by the explanation of the physical basis that makes that plausible. If it doesn’t match, then it’s dropped, again with some other physical explanation as to why it instead is sensitive to temperature, etc.
However, if these proxies are screened via dispassionate algorithm … Isn’t it possible that some tree out there can be said to be ‘sensitive to temperature’ some years, then not for others, all based on how it matches the actual temperature trendline? Further, couldn’t a tree hallowed for it’s sensitivity to temperature now (thinking Yamal), upon revisitation, end up being rejected for validity because of its divergence in later years? I wonder if there will be certain trees in-and-out repeatedly as a ‘valid proxy’ between now and 2050…
The time back is not so long, but grapes culd give an indication:
http://www.klimadebat.dk/forum/vindruer-kan-bruges-som-termometer-d12-e3297.php#post_60651
And: http://notrickszone.com/2017/03/06/historical-grape-harvest-dates-show-modern-temperatures-no-warmer-now-than-most-of-the-last-1000-years/#sthash.xJwz0uhF.6UzOpAc9.dpbs
They could especially tell how most of Europes temperature developed in modern time. And mostly growing season temperature where treerings tell an other story.
Interesting, thanks Svend. Here is the vegetative temperature proxy for the last millennia, yielding cogent results. This is the sort of data that refutes entirely the work of the dendros.
To historians the MWP is known as the High Middle Ages, a period generally put at circa 1000 AD -1300 AD. The historians usually attach no climatological significance to this period, which was characterized by unprecedented population expansion and general prosperity throughout Europe.
This population expansion and attendant prosperity was due to the longer growing season and consequent high crop yields. Put simply, more food meant more people and greater prosperity for all. I don’t believe that historians are cognizant of the effect of the warmth of that era, climate factors being generally outside their expertise. They generally attribute the subsequent decline in population, after 1300, to such calamities as the Black Plague. But we know that this decline coincided with the LIA, which probably was a greater calamity, as it severely curtailed food production.
Svend – This is one of my peeves with the dendros and other proxies. The proxy should coorrelate with the modern day instrumental record, subject to the obvious limitations, but also correlate/reconcile with other historical events. If not, then the calibration is likely off.
Is that a scientific reason to reject a serie?
“The following four PAGES 2014 Arctic sediment series were rejected in PAGES 2017:
the Iceberg Lake (Alaska) varve series (previously criticized at CA tag) because of its “unclear relation to temperature”
the truncated Korttajarvi X-ray density (XRD, a form of grey scale) see CA tag, originally used in Mann et al 2008 and Kaufman et al 2009, now also said to have an “unclear relation to temperature”
the previously truncated Igaliku series (CA tag)now removed in its entirety due to its “unclear relation to temperature”
the Lone Spruce (Alaska) BSi series (introduced in PAGES 2013, but only mentioned in passing in CA posts) was also removed due to its “unclear relation to temperature”.
How can they tell it is not responding to temperature?
You don’t know the temperature at that time, so what other parameters made them reject these series?
‘ “Cooking” also implies a malicious intent that is not applicable … there is a definite bias, but it’s mostly unconscious or rationalized with good intentions.’
That won’t wash, Mr McIntyre. I’ve always admired your self-control in not accusing obvious crooks of crookedness; after all, we can’t be certain of someone’s malevolent motivation, short of its being admitted in (say) leaked e-mails. It follows, however, that neither can we be certain that bias is ‘unconscious or rationalised with good intentions’.
Steve’s restraint is a rational decision. He points out that there is a definite bias. What he doesn’t do is ascribe motive, or if he does, then limits the seriousness of the motive to the minimum necessary. That is Occam’s Razor at work. There’s no need to ascribe to malice what can be explained by misinformed good intentions and expectations.
Duster’s comment is one of the great please-wear-blinkers recommendations of all climate science posts. He writes: “There’s no need to ascribe to malice what can be explained by misinformed good intentions and expectations.”
This exactly mirrors the analysis of the Ukrainian famine provided by the New York Times’ columnist Walter Duranty, whereby Stalin’s deliberate imposition of cannibalism on a great European nation was ascribed not to evil motives (as it should have been) but to “misinformation” about the benign motives of Soviet policy. After all, how could a government really deliberately cut off food from millions? The sad fact – that took until the Stalin-signed documents of the Perestroika revelations to emerge – is that Stalin got his jollies by torturing people.
Now just suppose that the benign view of the motivations of climate scientists is incorrect. Suppose that the ruined lives of dissenting scientists and journal editora are not due to any scientific errors on their part, but to organised conspiratorial malice akin to that directed against those (too few)who opposed Lysenko in the vain hope of saving Russian agriculture. Then choosing not to recognize the malice becomes akin to deliberate political blindness. How is it so much as possible that doctorally-degreed climate scientists are ignorant of proper criteria-set-before-the-hypothesis-testing, sampling techniques? When faced with a choice between hypotheses of genuinely innocent gross scientific incompetence on the one hand, and of an activist claque manipulating matters for their own political ends on the other hand, go for the claque hypothesis every time. People’s lives have been destroyed. Dismissing deliberate malice as the motivator for it, is to unwarrantedly rule out a possibly true hypothesis and thereby do the victims a very great injustice. Getting to the truth is hard enough, without putting up “do not investigate along this road” signs before the investigation has really begun.
macumazan: If anyone can be faulted for not bothering to investigate or giving up in the face of a highly motivated, highly funded and fearless cult creed opposition I sure it’s not Steve Mc. If one wants to help put sunshine on climate science (as I do) there is another Canadian who needs support right now for daring to infer ill of Mann’s scientific ethic. If his defense of his statement finally makes it to trial, which the individual is pushing for against legal advice, (after 5+ years) it could be the trial of the century, placing the question of knowing deception behind the MBH hockey stick under legal scrutiny. Steve has made all the keywords on this topic flag for moderation so I am trying to respect his wishes by directing discussion of the other site to the other site.
According to the report on the Hallet Lake, Alaska BSi proxy the summer air temperature increases exponentially with the SiO2 concentration. This means, according to this proxy that the early Holocene temperatures before 5000 BP must have been enormous! that is unless the very high BSi concentrations have to do more with the proximity of a huge glacier next to the lake earlier in the Holocene.
Glacier was more likely to have retreated in early Holocene and advanced in Neoglacial.
High BSi values in mid-Holocene occur in many other places and are evidence of warmth. There is considerable nonlinearity in some of these proxies – I do not believe that BSi increases “exponentially” with temperature – that would be very uncharacteristic. More likely some sort of S-curve.
I found that you had handled Hallet Lake BSi in an earlier CA post on Sept 22, 2009. Exponential only over the calibration period.
Amazing what one finds in old CA posts. The comments in that thread include a nice endorsement by Ronald Lanner, an authority on bristlecones, which supported the long-standing Climate Audit questioning of the bristlecone ring width chronologies relied upon in so many IPCC temperature reconstructions.
Steve:
Perhaps you recall the piece you wrote that analized the provenance and composition of some of the more widely cited tree ring series?
IIRC you created a really nice graphic representation of the interretationships between the data sets.
Ive tried to find it in your archives, and with some google searches but I seem to be missing the mark.
Can you help?
any more clues?
It was an analysis of the Dr Moreauesque combinations of various dendro and paleoclimatological proxies.
My recollection was that you took the time to trace the individual components of a lot of the proxy amalgamations.
The graphic was stunning…
The degree of interrelatedness was to me, shocking…
The same time series showed up in data set after data set; some data sets contained duplicated individuals, as the combinations of the individual sets sometimes contained data from same individual sample in two or more sets.
In any case what your analysis revealed, IIRC was that almost all of the statistical analysis was being conducted on the same base of data; most of of the data were merely different combinations of the same underlying constituent time series..
does this help?
Thanks for directing me to your old work, I went back and reread your take on Mannian PCA;
thanks again.
BTW: the “community’s” hostility to econometricians seems…
misplaced and unfortunate; after all its those people who actually encountered and learned to deal with the very problems the paleoclimatological community is struggling to comprehend now.
Of course Econometricians were doing that in the 1950s 1960s and 1970s-something like three generations ago.
It is to me cray cray (using the patois of the street), the astounding lack of knowledge displayed by some in that circle…
Like Fischer wasnt himself an economist?
I suppose Student was a poseur and a phony because he was a brew master for Guinness?
You are a much better and charitable man than I am.
@DE
Is this the post you were thinking of?
DaveS:
Thank you…while this post addresses the issue, I dont think its the one I was thinking about.
However it is good enough for me…
The graphic that sticks, with some vagueness, traced the constituent time series as the travelled through different combinations and recombinations….
Thanks again for finding this post, this is certainly grist for the mill.
I have followed Climate Audit for a long time and try to understand the issues discussed. If these lake sediments reflect temperature.do they represent an annual temperature or like my understanding of tree rings just the summer melt season temperature? Or to put it another way if one winter was say five degrees colder than the next winter would it show up in the sediments.
Nigel, its much the same as tree rings. The individual varves represent a year’s deposition and the thicker the varve the warmer the weather, supposedly. Same reliability as tree rings.
Nigel, here is McKay (2008) which studied Hallet Lake.
My question is why they are only interested in going back 2K years when they state the proxy record for the lake goes back to the mid Holocene?
The Hallet Lake chronology was not obtained by annual varve lines but by the trace markers of anthropogenic plutonium and lead marking sedimentation to 1963, which reported to indicate 0.76mm/yr for Pu and 0.82mm/yr according to Pb. The 4.5m core was determined to represent 4k years calculating to a 0.9mm/yr average rate. So McKay enjoyed the freedom here to squish and stretch the record by 10%-20% in order to fit local summer weather records (Valdez) back to 1917 to gain his correlation scoring.
Son Ron, the R-square of that regression is around 0.76.
That varve time series explains about three quarters of variance in temperature, for three months out of the year…
Im sure you are all too aware of the limited utility a proxy with these qualities.
Nigelj / Nigel NZ – welcome to climate audit where real science is actually debated. I noticed that you are a frequent poster at skeptical science (an activist site vs a scientific site) where any comment pointing out deficiencies, errors or limitations is strictly forbidden and will get you permanently banned. Your question is a good question which points to the limitations of the proxies used in temp reconstructions. The quality of the reconstruction is limited by the quality of the underlying proxies. With the frequent ex post selection criteria for proxies in most temp reconstructions, you should naturally question the scientific reasonableness of the conclusions. Hopefully some of the insight gained here will carryover to your posts at Skeptical science. (again I am presuming you are one and the same as nigelj from NZ who posts at SS – If you are not, then my apologies)
Not me, I think I looked at Skeptical Science once and have never been back, never posted there.
Just trying to get my head around what is going on. We have two types of proxies tree rings and lake sediments, both reject samples that do not have a correlation to the researchers perceived temperature record seemingly at every step in the process, those that do show a correlation it may or may not be related to temperature.
If they are related to temperature it is only a brief summer thaw period which can be less than three months and these are used to determine worldwide annual temperature changes.
My apologies for getting the wrong person. There is another nigel from NZ that regularly posts at Skep science. (one of most anti science websites imho). Again my apologies
from what I have read, varves are separated by a fine sediment layer deposited in a glacial lake during the non-melt (winter) season and a coarser layer deposited during the melt (summer) season. The thickness of the non-melt layer does not very greatly. The thickness of the melt layer tends to vary with climatic conditions.
The melt layer thickness depends on the velocity of currents in a glacial lake that are induced by the melting of snow in the area surrounding the glacier. Thus the thickness can depend on the amount of winter precipitation, the spring temperature and the distance from the varve of the glacier.
I hope that I have gotten this at least superficially correct.
TAG, I did a search in CA: “varves” a couple of years ago to find out what you are asking. I found it’s complicated. If I can recall, the glacier is receding from the proximity of the lake this would indicate warmer climate but thinner varves dues to sediment no making it to the lake. OTOH, warmer climate means stronger melt runnoff and more sediment. But the funny thing is that the climate scientists sometimes can’t decide themselves which varves are warmer, thicker or thinner.
Sometimes ambiguity is rooted in the indeterminate nature of a thing, which simply excludes their use as a proxy.
Lurked here many a time .. most interested in the ocean reconstructions .. anyone here know what happened to Oceans 2k ?
–
Their website said the 20th century bin would be published in May 2017 ?
Many thanks to CA and the contributors …..
Can I get a link please to the reworked post from 11UAG2010?
Original post: https://climateaudit.org/2010/08/11/within-group-and-between-group-variance/
“UPDATE Aug 13 – there was a mistake in the collation of trends that I used in this post -which was (confusingly) supposed to help clarify things rather than confuse things. I used trends calculated up to 2099 instead of 2009, which ended up making the within-group standard deviations too narrow. I’ll rework this and re-post. However, I’m going to Italy next Tuesday and am working on a presentation so the re-post will have to wait. This has also been a distraction from the MMH issues so let’s focus on them.”
[I would ask on that post but the comments section was closed. And there is no update on that old post linking to the new post.]
Also, it looks like some of the original post from 2010 was erased (based on the comments). Even it was in error, why not leave the post and allow the discussion of it to learn something. It will be hard to compare the new reworked post to the previous one otherwise.
Responding to the post, not the innumerable comments (many of which are OT).
It is incorrect to claim that PAGES2k discarded 50% of the lake sediment records.
PAGES 2013, v1.0 had 23 arctic lake records
PAGES 2013, v1.1., rejected 3 (see https://www.nature.com/ngeo/journal/v8/n12/full/ngeo2566.html)
PAGES 2017, v2.0, we rejected another 4 and added 3, for reasons explained in Table S2.
Werner et al CPD 2017 is a climate field reconstruction based on a slightly earlier version of this dataset.
They excluded non-annually resolved records for reasons made clear in the manuscript – there is nothing “strange” about that – unless you want to misconstrue it. The entire point of a compilation like PAGES is that it is relatively permissive, so users who are more stringent can raise the bar and use only a subset of records for their own purposes.
So, out of the original 23, 7 (30.43%) were rejected because of more stringent inclusion criteria, with 3 additions. Anyone is welcome to see what impact this made to an Arctic composite or reconstruction using a method that meets CA standard.
Finally, it is entirely incorrect to claim that PAGES 2k did not issue a corrigendum to identify the errors in v1.0 that were corrected in v1.1. They did so here (https://www.nature.com/ngeo/journal/v8/n12/full/ngeo2566.html), where Steve McIntyre was acknowledged about as clearly as could have been done: “The authors thank D. Divine, S. McIntyre and K. Seftigen, who helped improve the Arctic temperature reconstruction by finding errors in the data set.”
Continuing to whine about the lack of acknowledgement is beginning to sound like a delirium of persecution. We can certainly fix issues in the database, but Steve’s mental health issues are beyond PAGES’s scope.
Perhaps the CA tip jar pay for some therapy?
You too, seek help.
Following your own link I read that the Pages 2k Consortium only mentioned McIntyre first time in their corrigendum of November 27 2015, in between a few copycats. That is about 2 1/2 years after Steve showed some flaws here on CA.
Who needs therapy?
so, you use the pronoun “we” while commenting under some nom de net:
who are you?
Its not an unfair question to ask.
Probably some grad student repeating scuttlebutt and trash talk that he picked up in the corridors of USC.
“…It is incorrect to claim that PAGES2k discarded 50% of the lake sediment records out of the original 23, 7 (30.43%) were rejected because of more stringent inclusion criteria, with 3 additions…”
Nobody made that claim. Here’s what was stated:
‘The inventory flows of Arctic sediment proxies are summarized below. 32 “different” (non-isomorphic) series were introduced in the four studies as “temperature sensitive”. (For the purposes of this inventory, flipped versions are treated as one series.) 16 of the 32 series were rejected in a subsequent study as not being “temperature sensitive” after all.’
It isn’t just about “the original 23” from Pages2013. And it’s not just about being rejected by Pages 2013 and Pages 2017. It was even presented in a tabular format for people with simple reading and math skills to digest.
-snip-
The “whine” comment was uncalled for (it reads like a Julian Emile-Geay comment) as is the “grow up you pompous POS”.
Both detract from the substantive points that are being raised.
What would be helpful to me, and I suspect others, is a restatement of the differing views so we can see which proxies meet which criteria for inclusion in any reconstruction of Arctic temperatures.
Thanks, Bernie. Such gratuitous insults help nobody.
I disagree. I think that Jankowski presented useful information in his comment and exposed a falsity in the El Nino comment. This is much more enlightening than sanctimonious preaching, imo.
mpainter:
I also thought that Michael Jankowski’s comment was extremely useful. However, his final gratuitous comment undermines the need for El Nino or whoever to respond except in kind. The interchange illustrates Pielke’s point about Manichean Paranoia. I would rather we move the ball forward. If that strikes you as sanctimony, then so be it.
Bernie, moving the ball forward does not strike me as sanctimony. And I do not fret about El Nino types responding. They invariably spit and run.
Regarding paleoclimate reconstructions from lake varves, its the same old game of discarding the data that they don’t like and using the data that gives the right resultz, is it not obvious?
El Niño,
Could you highlight the phrases of this post in which Steve has complained about lack of acknowledgment? I reread the post and couldn’t find any.
Thanks.
Jeepers. First JEG, then whoever you are. Is it your _goal_ that onlookers should think of you-all at PAGES as a bunch of political partisans instead of scientists? Is it your goal to discredit your own work, by causing onlookers to think that winning a spitting match matters to you more than getting it right?
If that is your goal, you’re doing well.
El Nino is the handle used by Julien Emile-Geay (aka JEG) in his recent post about Pages2017 on the Strange Weather website. Unfortunately, he did not make this clear.
Hu:
Do I get a prize for recognizing his writing/commenting style? Didn’t Mosher have some kind of method for figuring out who wrote what during Climategate?
On October 7, 2014, I wrote the following email to Alicia Newton of Nature regarding McKay and Kaufman 2014 as a sort of “backdoor corrigendum” rather than an original publication. (Newton had been a co-presenter with me in 2006 at an AGU session on the NAS Hockey stick report.)
A few days later, I received a polite acknowledgement as follows:
On October 27, 2014, I followed up, additionally expressing concern about the fact that key changes in McKay and Kaufman had been plagiarized from Climate Audit:
I never heard back from them.
Until JEG brought it to my attention just now, I did not know that PAGES2K did issue a half-hearted corrigendum in November 2015 (!?!) over a year after my request to Nature. None of the authors had the courtesy to notify me or send me a copy.
This acknowledgement was not done voluntarily but only after I complained in writing to Nature and does not disprove my complaints about the earlier plagiarism by McKay and Kaufman 2014, the Kaufman et al 2009 Corrigendum and elsewhere.
I saw the notice of attribution to Steve for the Pages 2013 Nov 2015 corrigendum about a year ago and assumed he had been informed. At the time I said bully for them. Finding out that they did not announce it takes them back a big step.
On the other hand – I will note that skepticalscience is oblivious to the Nov 2015 corrigendum and its impact on their discussion of Pages 2013.
Hi El
from that Corrigendum this statement is made
“Following these corrections, the period from 1941–1970 emerges as the second warmest 30-year period in the Arctic record, and 1971–2000 the third warmest, rather than the first and second warmest as reported in the original version. The ranked order of the best estimate of temperature indicates that the warmest 30-year period is centred on AD 395. No major conclusions have been affected by the corrections made to the Arctic data set including the conclusion that, during the period AD 1971–2000, the area-weighted average reconstructed temperature among regions was higher than any other time in nearly 1,400 years.”
may be a dumb comment,but can you,or somebody at CA explain this reasoning/conclusion to me?
thanks
Dougie
Hi Doug, I puzzled briefly over that myself, but one needs to know paleoclimatespeak in order to make sense of such things.
Thanks for the reply mpainter
the clue is probably in “the area-weighted average reconstructed temperature among regions”
so they probably mean that since 1971 to 2000 the coverage of the Artic record/regions has improved & shows that the area-weighted average reconstructed best estimate temperature is the highest since AD 600ish.
only problem being they also give no.1 spot to AD 395 +/- something.
which misses the “nearly 1,400 years” for some reason !!
cheers
Very useful, that paleoclimatespeak. By such means you can obliterate the MWP and no one knows it. Poof! and it’s gone and there’s your reference for the next generation of climate studies.
I’ve added the following correction to the post:
.
Okay Steve….will you now please go seek psychiatric help?
I mean to be crazy enough to take on the religion of the left.
LOL just kidding….. ; )
You and your 5-year-old computer (at CA’s inception) put a billion dollar a year with all their super computers, watermelon fraud on its ear.
Muchos gracias muchacho, my grand kids don’t know it yet but will thank you……
Henry
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5226666/#!po=31.0976
This is a fairly recent article on the Lone Spruce lake proxy. Biogenic silicon (BSi) which acts as a proxy for diatom production, and thus, hopefully temperature seems to peak about 7000 BP after coming out of the ice age and slowly ramps down from then on. Plot is a bit noisy over the last 2 kyrs but nothing in the late Holocene has eclipsed the earlier readings ( Figure 3). Other variables seem to be unremarkable. This proxy was included in Pages 2013, but then disappeared in 2017 due to its uncertain temperature response. Might that be because it does not show unusual warming in the late 20th and early 21st centuries?
Biogenic silica, actually opal with about 5% H2O content, more or less. This a d18O proxy and I believe that this technique holds some promise. The opaline silica is from diatoms. Lone Spruce Lake is not fed by glacial meltwater but by a small catchment. The paleoclimate reconstructions from this study I consider reliable because of the proxy used.
This type of proxy might not work so well on a lake fed by glacial meltwater because ice at the bottom of the glacier is much older than that above, with the youngest ice at the surface. This age difference is a confounding factor and could corrupt the results.
The bottom line is that this study gives results that can be relied on. Solid science,imo. This superior technique should put an end, hopefully, to the garbage that has clogged the progress of paleoclimate studies.
So M…
whats the temporal resolution of this O18 proxy?
whats the precision?
whats its accuracy?
Not trying to be a difficult commenter, im really curious.
Hi, David. I’m no expert on d18O nor any of its applications, so I’m afraid that I can’t answer your questions. I know that it is by far the most useful of temperature proxies, giving fairly reliable results but it’s not infallible, especially with cave or coral studies and also Bsi, as I illustrated above.
BSi is an interesting proxy, especially on a Holocene scale, but has its limitations. I’ve got some interesting comparisons in inventory that I’ll try to write up this week.
Hi Steve, it should be interesting.
Please scratch the above comment. I misread the study and the “geochemical Bsi” does not mean an isotope analysis, so no d18O values were in this study but the Bsi was used as a biological proxy, for which I see little use.
D18O is so superior as a temperature proxy it seems inconceivable that the authors eschewed it. Bsi is of little use otherwise imo.
I did get to read the paper to which you linkied, and yes, BSi isnt an O18 proxy…
To clarify, Bsi has been used in a few recent studies for d18O analysis. The water contained in the silica makes this feasible, and excellent results can obtained. I expect this technique will eventually become the benchmark for other types of temperature proxies. Given the state of the science of paleoclimate reconstruction, this could take about twenty years or more.
So M…whats the temporal resolution on these types of proxies..
Centennial?
Decadal?
Multi- annual?
Annual?
sorry to pester you M. I will dive into it myself.
thanks for the time and thought you gave my question.
Again, I’m no expert but I would guess that resolution depends on the sample and that with adequate varve thickness decadal data could be achieved. But it’s a proxy and I would not put too much significance on squiggles.
David, in McKay (2008) the 5-year mean of BSi in Hallet Lake had the best fit to the instrumental station record of Valdez from 1917-2006. Remember, this is just the summer quarter of the year.
mpainter and Ron:
thanks…
chicken salad generally comes from chicken, not the type of data these proxies seem to be capable of yielding.
Steve,
Your table “Arctic Lake Proxy Inventory” has an error on the bottom row (CP17). It should read across:
2 5 7(not 4) 2 16
http://dro.dur.ac.uk/17008/1/17008.pdf?DDD14+dgg0jml+d700tmt
Have been looking at Holocene conditions in West Greenlandto see why the above diatom proxy may have been excluded. The main diatom proxy study was behind a paywall. This recent report is multiproxy from that area and all I can say is that for the last 3.5 kyrs it has been Frigid.
One Trackback
[…] https://climateaudit.org/2017/07/22/pages-2017-arctic-lake-sediments/#more-23301 […]