Von Storch reported that an early method for creating reconstructions produced reconstructions with reduced variability when applied to pseudoproxies containing noise. Do we have any idea if this new PaiCo methodology is capable of reconstructing the true dynamic range of climate variability?
[Response: The Von Storch claims were wildly overstated in the first place. See e.g. this piece in Science by Wahl et al. Subsequent reconstruction work using RegEM with TTLS regularization is quite resistant to losses of low-frequency variance. See e.g. the various papers by our group of the past 5+ years here. And of course, I discuss all of this in my book "The Hockey Stick and the Climate Wars". - mike]
Dr Kaufman: According to the Fig S1 in the supplementary information on the Pages website referred to above, the Pages 2K reconstructed temperature consistently overestimate global temperature in more recent decades by say 0.1 C. How has this been taken into account when concluding that “.. the average reconstructed temperature among all of the regions was likely higher than anytime in at least ~1400 years.” ? (sorry, but I can not read the paper from my current location).
Comment by Jens Raunsø Jensen — 22 Apr 2013 @ 5:31 AM
With respect to Frank’s question in #7: The PaiCo paper states:
When SNR is small, the noise dominates the pseudo-proxy records and, therefore, the low-frequency variability is underestimated while high-frequency variability is overestimated. This is expected from any method since, in high-noise cases, the noise ‘‘overwrites’’ the information about the target in the pseudo-proxies and, therefore, no method can recover the low-frequency variability. When there is little noise, the difference in power spectrums is much smaller. It seems that the error of PaiCo can be decomposed to a slight overestimation of the millennial to centennial scale variability and a slight underestimation of the centennial to decadal scale variability. However, as shown in Fig. 4, the errors in the reconstructions are among the smallest of any reconstruction method in many of the tested settings.
Nick Stokes has done a real service with his “Active Viewer” gadgets, most recently for the PAGES-2k datasets. As someone trained as a physical scientist who has taken lots and lots of data in my time, I can’t help but be struck by the variability of each and every one of these proxies.
To my eye, flipping between the proxies from a given region shows there to be lots of noise both within a given series and between any pair of series. This is very evident for the various ‘CAN composite’ series, which from Nick’s map appear to be geographically close together. When I spend some time flipping between these it is really impossible for me to say with any confidence that they are measures of the same thing – whatever that thing might be – and that if they are then the signal to noise ratio is miniscule.
If I had a student who brought data like this to me along with a statistical analysis I would send him back to the lab to get better data. Paleoclimate studies can’t do this, of course, but I’m wondering if it would at least be worthwhile to calculate correlations between proxies absent any underlying assumptions in order to assess whether the data has any value.
For example, wouldn’t one expect all the geographically close ‘CAN composite’ series to be more similar to each other than they are to, say central Asia? Shouldn’t it be a requirement that this be true before going ahead and throwing it into the mix? Otherwise it seems to me that garbage is being mixed with what is already a very small signal and I can’t see any good coming of that.
If I had to give an overall impression of the state of progress of paleoclimate reconstructions over the past 15 years I would say that it largely consists of taking more and more data and dumping it into ever more elaborate statistical procedures in the hope that a silk purse will emerge. It’s not at all clear to me that we wouldn’t be better off going back a step and trying to improve the data.
[Response: And what specific suggestions for "improving the data" do you have in mind? It is worth noting that in my experience, the correlation among weather station data is no higher than among similarly spaced paleoclimate data. To me, this says that the paleoclimate data quality if just fine-- and that the noise is in regional vs global climate, not the quality of the proxies themselves. In short, we need more data, not "better" data (though better is fine, obviously). Getting more data, of course, it exacty what the authors of the PAGES 2k paper are trying to do.--eric]
Darrell Kaufman says ‘For example, there were no globally synchronous multi-decadal warm or cold intervals that define a worldwide Medieval Warm Period or Little Ice Age’.
To a naive observer, this isn’t visually obvious. The dominant color at the year 1000 is at the warm end of the spectrum. The dominant color at the year 1650 is very much at the cold end. The temperature anomalies may not be global, in Dr Kaufman’s words, but certainly look very widespread indeed.
Linking’s a bit off still
— for readers if you don’t get one to work,
compare what you see hovering the mouse pointer over a link,
what you see hilighting the text containing the link and using “view source” in your browser.
Some links in text jump to footnotes at bottom of main post, which is fine.
Some links that are meant to go to another web site get “realclimate” stuck on the beginning, so although the correct URL is included it isn’t working.
You seem to have fallen into the same trap as the denialists–thinking that “close” counts as simultaneous, or that “a little warm” is as good as much warmer. Yes, it is true that there were several warm areas around the end of 1st millennium CE. However, the trend was not global, and the warming was nowhere near as strong as what we have seen recently.
“… The most coherent feature … is a long-term cooling trend, which ended late in the nineteenth century…. There were no globally synchronous multi-decadal warm or cold intervals that define a worldwide Medieval Warm Period or Little Ice Age ….”
But — lacking access to the full article text — I can’t be sure what those various colors represent. Does the same color for each proxy over the whole time span mean
– the actual temperature at that location (however interpreted)?
– the anomaly from a baseline for that proxy in that location, up to that time?
It’s probably easy to understand once you understand it, but it can be hard to get a picture that -by-itself- conveys what’s known.
I’d love to see what Robert Rhode would do with that if he gets globalwarmingart updated.
Was there not a peak global average temp around CE 1000, as reconstructed, equal or a little warmer than today?
I thought the issue we have now is that we have sudden, unnatural warming and are headed to temperatures higher than those in the last 10K years due to total GHGs and cascade effects (such as darker poles due to melting ice).
I’ve seen numerous charts of historical temps and find the comparison of today with yrs around 1000 not entirely clear.
Look at the pictures there at least — look at
Figure S2 Proxy temperature reconstructions for the seven regions of the PAGES 2k Network.
Looking at that, it appears that yes, there was not a peak where you thought you heard about one.
Where did you hear the idea that such existed? There are local ones like the one in Europe most people know about, but that’s not global.
[Response: This is a common misconception! I have never seen any evidence to support a "global" MWP, but you hear it all the time, and it is definitely taken as fact in much of the older literature. But that was in the absence of evidence. We no have lots of evidence and yet no global MWP, the PAGES 2K paper just being one of the more recent looks at this. --eric]
By improving the data I’m suggest attempting to screen out proxies which only (or mainly) contribute noise. Take for example S. America CAN Composite 11 and 31. An admittedly quick and “unscientific” visual comparison indicates a poor correlation between them. Yet they are separated by only roughly 1 degree. That’s roughly the distance between, say, New York and Washington, D.C. Now, clearly the weather in NYC and WDC is not going to be the same, but I would venture a guess that a cold year in NYC would more often be accompanied by a cold year in WDC than a cold year in, say, Moscow.
My thought was that if one knew what to expect then one might obtain a measure of the quality of the proxy series, and that having a number of closely spaced proxies might allow one to screen out data a priori which hurt rather than helped the analysis.
I guess I should frame it as a question: has anyone ever looked at something like this? Specifically, how does the correlation between proxies spaced by X km compare with the correlation between weather stations spaced by X km? Control for meteorological factors such as altitude, proximity to oceans, etc. would obviously have to be considered.
“has anyone ever looked at something like this? Specifically, how does the correlation between proxies spaced by X km compare with the correlation between weather stations spaced by X km? Control for meteorological factors such as altitude, proximity to oceans, etc. would obviously have to be considered.”
Since you are a self-proclaimed physical scientist, why aren’t you searching the literature yourself, rather thn trying to cast doubt without bothering?
The persistent idea that there was a “Big MWP”, i.e., a global MWP warmer than today was essentially manufactured in 2005 by wide propagation of Fig 7.1(c) from IPCC(1990). Basically, it was falsehood, but well publicized. For the history, see Wiki Talk page. This really got going in March 2005, helped a few months by the Wall Street Journal.
I’ve found at least 7 versions of the IPCC(1990) graph on p.202, none of which are the exact image, none of which showed any idea of the caveats around it, none of which admitted that the graph was heavily deprecated within a few years as unusable, and some instances of which claimed it was from 1995, in 3 instances specifically claiming ti was Figure 22, which in nonexistent.
The first link on the word “published” points to this page (to #ITEM-15406-0).
[Response: Clicking on that will point you to the full reference at the bottom of the post, which then has a link to the published paper here --eric]
On MWP: there’s a contrarian site whose URL I forget (co2science is in the name so you can find it easily) that has amassed papers purporting to show that there is a conspiracy to suppress the MWP. That they found so many papers makes for an odd suppression conspiracy. I selected the papers they claimed had the most reliable data sets, and the dates don’t line up. So it’s not a huge surprise that someone doing good science found the same thing.
If contrarians really were sceptics, they would pick this sort of thing up themselves rather than go to enormous lengths to accumulate data that undermines their case, then persist with their claims.
“Since you are a self-proclaimed physical scientist, why aren’t you searching the literature yourself, rather thn trying to cast doubt without bothering?”
Quite simply because I don’t get paid for it and don’t have the inclination to do so.
If there weren’t a problem with signal to noise ratio in paleo data then the extraordinary statistical efforts in each and every study I’ve looked at would be unnecessary. If this sort of study hasn’t been done already I think it would be effort well spent for someone in the field, probably worth a paper or two in its own right, and should contribute to improving confidence in the resulting reconstructions.
In my field I have the luxury of being able to integrate experiments for 24 or even 48 hours to overcome poor SNR’s, but do so only as a last resort after trying everything I can think of to get a “clean” signal. All I’m doing here is suggesting something that occurred to me that might help approach that goal in paleo studies.
As for casting doubt: even though that was not my intent in this case I must insist that is the essence of a scientist’s job. No theory has ever been proved “right”. The best we can do is to come up with hypotheses which don’t appear to be wrong. If we — or better still others — work hard enough to disprove them and still can’t, then we get to call them theories.
The problem is that if you use only “clean” data, you cannot say very much. It’s been done. Long ago. We are now in a situation where we are looking back thousands of years–those datasets will of necessity be noisy. That does not, however, make them worthless. They still contain information–signal–and it is information we cannot get any other way.
As we say in physics–all the easy problems have been solved, and they were solved before they were easy.
A question (or a few) about the ‘Sign relation’ attribute in the proxy metadata: This is presumably supposed to represent the relationship between the proxy data and temperature, such that a negative sign relation would mean higher proxy values indicate lower temperature?
If that is the case, has this sign relation been pre-applied to standarise the proxy data stored in the spreadsheets? I ask in relation to the Canadian tree-ring proxies, having been pointed in that direction by Watcher, across which there is a robust low signal in the early-1800s, coincident with the 1815 Tambora eruption. However, some of these proxies are listed with a negative sign relation. Does that mean the low will produce a warm spike rather than a cold spike when calibrated for temperature? Or am I barking up the wrong… well, you know?
Ray Bradley’s Paleoclimatology (1999) is still pretty useful for an introduction to the challenges of paleoclimate reconstructions and ways of dealing with them. (That’s the book plagiarized and then falsified in the Wegman Report in 2006.)
I’m lucky to have known or worked with many *good* scientists, most of whom refrain from disparaging an unfamiliar field before first learning enough about it for their comments to be relevant. Anonymous Dunning-Kruger Effect is alive and well.
“The problem is that if you use only “clean” data, you cannot say very much. It’s been done. Long ago.”
Well, I can’t say I’ve ever seen a paleo study that uses what I would call “clean” data, though I admit to being a dilettante in the area. What was the source of the data and how was it known to be clean?
Getting back to my original point, which I should probably soften. Rather than checking whether the data “has any value”, maybe I should say it might give an estimate of how much value it does have. Because the discussions here have forced me to think this through a bit (funny how that works!): if one takes several years worth of data from weather stations spaced some distance apart, some measure of correlation between their annual average temp can be calculated, the more years the better. A correlation can also be calculated from proxies located a similar distance apart. In the best case they would be the same number, but in reality the proxies will almost certainly be worse. Just how much worse should be a measure of the noise in their signal. Have there been studies like this?
Perhaps close-proxy correlations could be used to generate weighting factors when used in a reconstruction, or perhaps to provide a threshold criterion for whether they should be included.
Principal component analysis and the other techniques do basically what you are talking about. Remember, though, you are doing a reconstruction over thousands of years. Sometimes the reason a problem is hard is because it is inherently difficult.
And there is a big difference between asking dumb questions to famailiarize yourself with a new field and implying those in the field don’t know what they are talking about. In the words of St. Patrick, “Oh,Lord, let my words be sweet and gentle today, for tomorrow I may have to eat them.”
If I’m not mistaken, most of the paleo datasets have multiple samples taken in close proximity to each other (spatially). Marine samples immediately come to mind. In that way, you get some idea of the statistical uncertainty for each proxy location. In the case of EPICA, two ice cores were eventually obtained.
Spatial autocorrelation works best at the high end of the frequency spectrum (hours, days, months or years), but most (if not all) paleo data is essentially a low pass filter, since the measurements are not instantaneous by the very nature of proxy deposition (e. g. air pocket close out depth for ice cores (snow -> firn -> ice), diffusion, etc.).
It’s not like those doing paleo studies have overlooked the obvious (e. g. are the proxy data similar to each other at the locations where those specific samples were taken).
OK, at great personal risk because there seem to be some thin skins around here….
I originally referred specifically to datasets taken in close proximity, which to my inexpert eye did not look very similar at all. Dr, Steig responded with
“in my experience, the correlation among weather station data is no higher than among similarly spaced paleoclimate data”
In retrospect I suppose this could mean that a) neither proxies nor weather stations correlate well, or b) that they both do. Given my data pair example and the way he phrased it I took his comment to indicate the former sense is what was meant.
I find it surprising that closely spaced weather stations do not correlate well. I frequently do short-hop travel between cities with roughly a 1 degree lat/lon spacing, and I find the correlation pretty good as long as a day or two is allowed for weather systems to travel. A cursory scan of the weather channel leads me to the same conclusion. I would expect the effects of a few days lag would disappear over the annual or greater time scales of proxy data.
I’m sorry if I sound pig-headed and acknowledge that my anecdotal experience can hardly be considered definitive. If it is the case that this issue has been dealt with in the literature, then perhaps someone could point me to an appropriate reference and we’ll call it a day. If not, I still think it would be great use of a grad student’s time to at least explore the area.
I actually have repeatedly asked pseudoskeptics pointing to the CO2science website to make such a graph. Usually silence ensues, because they realize that suppressed knowledge can’t be hiding in plain sight.
Contrarians find it difficult to argue against AGW when faced with even a single hockey stick, which is why they try so hard to break any they come across. One attack-route is to harness the good old IPCC FAR figure 7.1c as their evidence and to claim the LIA & MWP have been wrongly airbrushed from the record. Thus it can be argued that contrarians began ‘manufacturing’ the MWP after the 2001 IPCC TAR gave the Mann et al 1999 graphic widespread publicity. The use of fig 7.1c is usually pretty childish stuff, as this debunking of William Happer demonstrates.
It looks like the BEST folks had a look at correlations between weather stations. A summary graph can be found on page 17 of BEST Methods paper’s appendix. Whether the times scales they used would be appropriate for a proxy evaluation I can’t say, but they show correlations of 0.5 or so for 1000km separations.
If proxies 1000km apart are perfect thermometers we should expect a correlation similar to weather stations as calculated on an appropriate timescale. Unless I’m missing something this offers an independent means of assessing the quality of closely grouped proxies.
“If proxies 1000km apart are …” at different elevations, or rural vs urban, or on hilltops vs small valleys …. one wouldn’t expect perfect correlation [thermometers or proxies], even if on the same latitude line. What exactly are you questioning, Watcher, the scientists studying temperature records and reconstructions have a pretty good handle on the complexities of their field.
Watcher- are you talking about correlation in the anomalies (differences from baseline) or the actual temperatures… teleconnection (the expected correlation of stations to each other in space) to my understanding is more for anomalies, not for temperatures.
Quite correct, which is way back in #20 I mentioned “Control for meteorological factors such as altitude, proximity to oceans, etc. would obviously have to be considered”. In the BEST plot they say they used random pairs of stations, which is probably why the scatter is so large. I would venture a guess that if one did correct for conditions the correlation would be even stronger.
I mean, think about it: when you check the weather on TV they show the fronts moving one way or another and they know if it will be warmer or colder tomorrow at your location by what happened in the neighboring state today.
“more for anomalies, not for temperatures”
Correlation between stations, period. Weather stations measure temperature, not anomalies (i.e. we’re tossing out all the data on humidity, pressure, etc) so that’s what you need to assess.
Earth’s land surface temperature trends:
A new approach confirms previous results
Barbara Goss Levi
Physics Today / Volume 66 / Issue 4 / April 2013, page 17 http://dx.doi.org/10.1063/PT.3.1936
“… The newcomers to the task looked at many more weather stations and used a geostatistics technique to adjust for data discontinuities….”
“… write the temperature measurement for a given place and time as the sum of four terms: an average global temperature Tavg; the positional variation caused by latitude or elevation; the measurement bias, or offset variable; and the temperature associated with local weather….”
“Figure 1: Mean correlation versus distance curve constructed from 500,000 pair-wise comparisons of station temperature records. Each station pair was selected at random, and the measured correlation was calculated after removing seasonality and with the requirement that they have at least 10 years of overlapping data.”
Seems clear enough. I don’t know how to post figures so the best I can do is direct you to page 17 of the link in #40.
I’ve made no comment at all about their reconstruction, which has nothing to do with proxies. I simply happened across their station correlation data which I presented as support for my hypothesis that closely spaced weather stations should be better correlated than more distant ones.
That is the info I needed, especially the link to popularized form on SkS. I can’t follow a full paper, due to in some part to time but mostly to health limitations. I like to know where to find it tho.
The notion of a medieval warm period is so common I thought it was generally accepted. The term is used on Skeptical Science in the link but I could have picked it up from elsewhere. I’ll hazard a guess you could find it in Scientific American if you go back a ways.
I’m having trouble reading the graphs in that link though: in Mann 2008 it looks like a 2/10th excursion around yrs 800 and 1000 in the later PAGES it looks smaller, about .1 c and only around yr 800. I am not sure which figure to use on the next time someone tells me “herr, it wuz hotter when the Vikings xxx”
Actually, the denialists are playing to prejudices of those of European descent. There was definitely a warm period in Europe in this era, and since European history is all most people will have any familiarity with–even most historians–it is easy to get people to view it as global.
Denialists always hope our native stupidity is transferable to the particular stupid thing they want us to believe.
In response to my original suggestion of calculating the correlation between nearby proxies as a cross-check on their quality Dr. Steig responded with,
“in my experience, the correlation among weather station data is no higher than among similarly spaced paleoclimate data”.
As I mentioned earlier (#35) I had taken this to mean that neither weather stations nor proxies were very well cross-correlated, but in light of the BEST team’s data I guess I need to revise that. Taken as written it implies that proxies should have a higher degree of cross-correlation than weather stations, but I doubt that was the intent and will assume it to mean they should be the same.
Given the discussions above of geographic factors in modifying the correlation of weather stations, it also occurred to me that there must be decades of temperature data from commercial weather reporting for many of the areas where proxies originate. Accordingly, I feel justified in asking again: could these data be used to come up with a “target” cross-correlation fro the proxies? And could these cross-correlations form a basis for assessing the signal to noise ratios in the proxies?
I’m really having trouble understanding why this idea seems to be dismissed out of hand.
[Response:I didn't mean to appear to have dismissed anything out of hand. I was merely responding to a specific assumption you were making, which may not always be correct. I think your suggestion about cross-correlations for for assessing the signal to noise ratios in the proxies is good. --eric]
> my hypothesis that closely spaced weather stations
> should be better correlated than more distant ones
But they say they make the adjustments they describe.
“the temperature measurement for a given place and time as the sum of four terms: an average global temperature Tavg; the positional variation caused by latitude or elevation; the measurement bias, or offset variable; and the temperature associated with local weather”
They adjust away the latter three terms, leaving the first one, right?
(Someone will surely correct this oversimpification)
Seems like you assume the caption for a figure is the complete statement of the adjustment, then your hypothesis — derived from reading that one caption — is that the paper looks wrong.
Could be your reading of that caption is incomplete.
Just sayin’, doubt yourself first.
They’re nice folks over at BEST, they do have a FAQ and publish their work. You might do better asking them there.
The thin skin appears to be mostly yours. It seems relatively straightforward that paleo data are difficult, and if you assume good faith on the part of the majority of the scientific community, along with relative intelligence, you might not be so ready to take offense when you suggest that they do their jobs, and that they have not already done so. Because the temperature record is important and has been under constant attack, well supported by industry which has vast funds to put into an alternative universe of think tanks and detractors, scientists have strained every muscle and every brain cell to come up with ways to measure, be honest about uncertainty, qualify it, and extend the record.
This effort is worthwhile, but defending it is wearying, as the same old suspicions pop up hundreds and thousands of times, unchanging and relentless and unaccepting of on-topic responses. You may not be aware that you are treading a well-trodden path, but your first statements looked familiar.
We are never going to get perfect data for thousands of years ago, but this is a worthy effort. If we start with the premise “what can we learn from this” instead of “what’s wrong with it” we could get on.
Perhaps you are genuinely interested in learning more, but your approach was guaranteed to raise the hackles of a community under constant attack from forces that are consistently hostile and prone to use what they know about science to destroy rather than build knowledge (and these attacks are not limited to science, it gets personal when your family gets physical threats).
It seems obvious that when data sets are different by necessity, but measure the same thing (temperature over time), it will be easy to attack the correlations by saying they’re not perfect, but life isn’t perfect. The suggestion that different measurements of temperature should not be compared is silly if you think about it. The best we have is to find ways to use them together, with care and honesty to qualify the ways in which they diverge from what we’ve been able to learn. We’re stuck with it, on our only planet.
One can hope you came here because you realized that some of the world’s top climate scientists are freely sharing with the rest of us in an effort to improve communication. If opening questions sound like a suspicious policeman with an agenda, it’s not surprising that others more familiar with the subject matter than you are call out your assumptions.
Comment by Susan Anderson — 27 Apr 2013 @ 10:25 AM
OK, one more time. Just because I don’t know when to stop.
Hank Roberts (53 etc) still seems to think I’m criticising the BEST reconstruction, when all I’ve done is take one of their background analyses and used it to make AN ENTIRELY DIFFERENT POINT which Dr. Steig has been gracious enough to concede may have some merit.
Susan Anderson (54) makes a number of points which I would like to address.
Trust me, my skin is pretty thick and my somewhat repetitive postings have been born out of a frustration at not being understood. I suppose I need to accept some responsibility for not expressing myself clearly, but in at least some cases (Hank shall remain nameless here) folks take their preconceptions about what they are expecting and react to those.
As for suggesting folks aren’t doing their jobs: I mentioned that I was trained as a scientist, and in fact I continue to work as such — though more in the engineering line these days. I’m constantly involved in technical discussions which I treat very differently from social discussions. In a technical discussion, if you think you have a valid point, you worry that bone until you get your point across. It has nothing to do with motivations or other peoples’ feelings, because in the end Nature is the arbiter and it doesn’t matter who THINKS they’re right.
If you review my allegedly hackle-raising approach, you’ll see that at no time did I impute any motives or threaten any families. I gave a technical assessment (qualitative, admittedly) of some data. I gave a technical suggestion on how it could be weighted during analysis. Since this is a blog about scientific discussion I felt it an appropriate way to proceed, and still do.
At a dinner party I wouldn’t tell the host their cake would be better with butter instead of shortening. I’d just offer to bring dessert next time.
Please continue. I’m not a scientist, just another reader here. I’ve been trying to figure out what you’re trying to ask, without success; I haven’t figured out what the assumption is that you’re starting from. That’s just me, most likely.
Someone else will be able to understand you, if you persist.
Best bet — ask your question in a way that will attract the interest of one of the scientists here; they usually notice and reply rather quickly once you ask a question that’s clear enough to answer. (And when they do, folks like me learn from both question and answer. Seriously, do persist.)
I’d bet that the analysis you propose has been done many times by many people, but it has an issue with the “hide the incline”* that contrarians point out – that the instrument record has a larger increase than some proxies would suggest. Thus, I’m not sure success is even possible until we figure out why there has been a divergence with some proxies during the modern era.
*of course, contrarians use the word “decline”, which is ludicrous. The thermometer record has INCREASED, not decreased as compared to some proxies.
I’m afraid I have to disagree with much of what you said.
If it has been done many times then it hasn’t been published, at least not within the last few years since I’ve been paying attention to this stuff. As was pointed out by another commenter there are professional climate guys here and I’m confident that if I was missing some existing body of literature on the subject it would have been pointed out.
It has nothing directly to do with “hide the decline” since the question being asked is very simply “are proxies correlated with each other to the same degree that thermometers are correlated with each other”. Either or both can be moving up, down or sideways without prejudicing the result of the correlation. It doesn’t require that the correlations be calculated over the same times frames. It doesn’t even require that the sign or magnitude of the temp/proxy connection remain constant*. If two weather stations read similar temperatures over the course of last year the odds are good they did so last century as well. Or last millennium. When the correlations between a corresponding pair of proxies is calculated no assumptions about calibration to temperature are required. It thus avoids the criticism of “pre-selection” of proxies based on the short overlap period with the instrumental record.
* I’m on thinner ice here, but I think that’s true!
[Response: Of course, it's not simply "how well are they correlated", because that depends on the timescale. One could have proxies that are poorly correlated with each other on annual timescales, but well correlated on longer timescales. The physics determining the long term variability could be related to the temperature, but the physics pertaining to the short term variability might not. Actually, a reverse sort of example of this would be raw tree ring data, where the very lowest frequencies are determined by the age of the tree (and hence the need for multiple trees of different age, corrections for the growth curve, etc. etc.).--eric]
If I might paraphrase your response: “The devil is in the details”. Quite right.
I can imagine a pair of thermometers, one at each edge of Oklahoma, say. Having driven across Oklahoma one can’t help but be struck by the OMG sameness of the place, so overall one would expect a pretty tight correlation between the two stations. Further imagine a cold front passing West to East. The temperature variation would have a time lag so the correlation would be there, but “out of phase” if you will, using instantaneous temperature. If you used a daily max/min you’d probably reduce that effect. With weekly or annual averages maybe get rid of it altogether.
So yes, I agree you’d need to examine whether it’s better to use monthly, annual, etc. data to correlate with the (at best) annual data from proxies. Lots of tedious work, but then that’s what grad students are for, isn’t it?
[Response: ;) re grad students. An example of what you’re talking about is my (at the time) grad student’s work on Mt. Logan. The “proxy” is annual snow accumulation rate, the climate target atmospheric circulation. Winter is the best time period. The proxy works just as well (or better) than the instrumental precipitation data. http://faculty.washington.edu/steig/papers/recent/RupperSteigRoe.pdf,
Thanks for the link. The paper is not something I’m going to pretend to have digested in a quick scan, but what’s clear right away is that even ignoring the two high altitude locations (Mt. Logan and Wolverine Glacier) correlation patterns for instrumental precipitation are pretty complex. One hopes that overall temps would be simpler or the method wouldn’t be of much use.
Probably none. It’s just a “thought experiment”. It’s often useful to consider the most idealized case you can come up with before layering on the complications that arise in the real world. And before anyone piles in: yes, Oklahoma exists in the real world.
In fact, Oklahoma (or any other of the Plains states) is a bad example of a place where you’d expect tight correlation. The east-west gradients are strong and are related mostly to soil moisture. Annual rainfall ranges from >50 in. in the east to <20 in the Panhandle. Differences in the annual temperature are almost 8 F (warmer on average in the south and east), but day to day differences in the spring are typically reversed from the annual averages. Today, for instance, highs in the west were low-to-mid 90s, with the east in the low-to-mid 80s. Vegetation in the east is typical of eastern deciduous forests and, in the west, the scrub vegetation associated with the high desert.