The subject of reconstructions of temperature variations of the past millennium has been discussed many times before on this site (see e.g. here, here, here, and here). Despite the apparent controversy, the basic conclusion–that the global and hemispheric-scale warmth of the past few decades appears anomalous in a very long-term context–has stood up remarkably well in many independent studies (see Figure 1).
Figure 1. Reconstructions of Northern Hemisphere temperatures for the last 1000 years (various colored curves) compared with instrumental record (black curve).
[source: Wikipedia] (click to enlarge)
This is not to say that all estimates agree in their details. Indeed, there is a fair scatter among the various published estimates. Some of these differences are believed to reflect differences in seasonality and spatial emphasis. Past summer, extratropical temperature changes appear, for example, to have have differed significantly from annual temperature changes over the entire (tropical and extratropical) Northern Hemisphere, and tropical Pacific Sea Surface Temperatures appear to have varied oppositely with temperatures in the extratropical regions of the globe. See for example the review paper by Jones and Mann (2004), in particular the discussion and references in section 5.3 therein. Some differences appear to be related to the particular method used to “calibrate” the proxy data against modern instrumental records. While the various methods can be tested with climate model simulations , it would arguably be more satisfying if inferences could be obtained in a manner which bypasses the difficult issue of calibration entirely, and also eliminates any need to establish the precise seasonality of information reflected by the various available proxy records.
This is what Osborn and Briffa have done in their article “The Spatial Extent of 20th Century Warmth in the Context of the Past 1200 Years”, which appears in the Feb 10 issue of the journal Science. The article uses a rigorous statistical methodology to re-examine the question of whether late 20th century warmth is anomalous in the context of the past 1200 years. This is done in a manner that does not require the explicit calibration of the proxy records. In essence, the authors have revisited a question posed earlier in a paper by Willie Soon and Sallie Baliunas (2003: see our previous discussion here), investigating whether or not evidence from past proxy records of temperature support the existence of past intervals of warmth with the widespread global scale of 20th century warming. The Soon and Baliunas (2003) paper was heavily criticized in the scientific literature (e.g. Mann et al, 2003) for failing to distinguish between proxy evidence of temperature and drought or precipitation, and for not accounting for whether temperature anomalies in different regions were contemporaneous or not.
Osborn and Briffa, by contrast, have carefully taken these issues into account. They make use only of those proxy records which demonstrate a statistically significant relationship with modern instrumental temperature records, and which were dated accurately enough that records from different locations could be compared against each other in a chronologically consistent manner. They then standardize the records and look for evidence of simultaneous relative departures that point in the same direction (i.e. “warm” or “cold”) using appropriate pre-set thresholds for defining a significant event (they try both one and two standard deviations). There is an important distinction between this careful statistical approach, and the selective cherry picking that is often used by contrarian commentators to misrepresent the available evidence. For example, it is possible to find evidence of significant warmth or significant coldness over literally any century-long interval in at least one of the 14 records used by Osborn and Briffa (see Figure 1 in the article). However, this alone tells us very little. What is of interest, instead, is whether centuries-long intervals can be found over which warm events or cold events tend to cluster.
Osborn and Briffa use Monte Carlo simulations to test the null hypothesis that a given number of simultaneous “warm” or “cold” events should simultaneously emerge among 14 such independent records from chance alone. Where they are able to reject this null hypothesis, they conclude that there is evidence of large-scale warmth or coldness, and they quantify its spatial extent. They also establish that their results are robust with respect to the elimination of any single record, and that the main conclusions are independent of the particular base period (e.g. whether they use the full interval AD 800-1995, or only the modern interval of 1856-1995 which overlaps with the instrumental record). While some of the caveats of some past studies are applicable in this study too (for example, the authors only use 14 proxy sites), the authors do attempt to address them. For example, they show that the modern instrumental record averaged only over their 14 sites captures the full Northern Hemisphere mean temperature variations remarkably well over the available (approximately 150 years) interval. The authors also examine the difference between the number of significant warm and cold events over time, and this tells a similar story.
Figure 2 (from Osborn and Briffa, ’06). Fraction of the records available in each year that have normalized values > 0 (red line), > 1 (light red shading), > 2 (dark red shading), < 0 (blue line), < –1 (light blue shading), and < –2 (dark blue shading), with the latter three series multiplied by –1 before plotting. The series are shown from 800 to 1995 and have been filtered to remove variations on time scales less than 20 years. [source: AAAS] (click to enlarge)
In each case, the authors find that the most widespread warmth by far is evident during the mid and late 20th century. The conclusion is not especially surprising, as nearly all previous peer-reviewed studies over the past decade find that late 20th century warmth is anomalous in a millennial or longer context. Indeed, the curve they produce (Figure 2)–with a modest negative trend over most of the past millennium ending in a dramatic positive 20th century spike–might be likened in shape to a certain implement used in a popular North American winter sport. But we digress…
It is not so much the conclusion, but the approach that the authors use to reach their conclusion, that is most important about this latest study. The authors take advantage of a very straightforward analysis of climate proxy data, avoiding the highly technical and arcane issues of statistical calibration and methodology that are so frequently seized upon by those who dispute that recent large-scale warmth is anomalous in a long-term context. This paper adds to the mounting weight of evidence that such warmth is indeed anomalous in at least a millennial context. We doubt that this, or for that matter, any study will silence the increasingly small but persistently vocal minority of contrarians who continue to challenge this conclusion. But to them, we offer the reminder that paleoclimate evidence comprises only one of many independent lines of evidence indicating a primary role of human activity in modern climate change. If the only line of evidence that remains in dispute pertains to estimated millennial temperature histories, then the case for denialism appears extremely weak indeed.
Jones, P.D. and Mann, M.E., Climate Over Past Millennia, Reviews of Geophysics, 42, RG2002, doi: 10.1029/2003RG000143, 2004.
Mann, M.E., Ammann, C.M., Bradley, R.S., Briffa, K.R., Crowley, T.J., Hughes, M.K., Jones, P.D., Oppenheimer, M., Osborn, T.J., Overpeck, J.T., Rutherford, S., Trenberth, K.E., Wigley, T.M.L., On Past Temperatures and Anomalous Late 20th Century Warmth, Eos, 84, 256-258, 2003.
Soon, W. and Baliunas, S. Proxy climatic and environmental changes over the past 1000 years, Climate Research, 23, 89-110, 2003.
54 Responses to "A New Take on an Old Millennium"
Interesting. Of the 10 reconstructions, 6 of them include authors listed together in your second reference (at the bottom of your article), while a 7th includes a further author (Schweingruber) who is also a joint author of one of the others. Osborn and Briffa are also included. So most of these reconstructions were carried out by a group of interconnected scientists. This doesn’t sit well with the claim of “many independent studies”. Just an observation.
[Response: This isn’t the place to play Six Degrees of Kevin Bacon. Lets keep this on the science from now on, and lets avoid ad hominems. – mike]
Hans Erren says
could you please explain why you included bristlecone pine series and the gaspe data set in your proxies?
[Response:They’re not ‘our’ proxies, but judging from the paper and supplemental info O&B only selected proxies that passed a screening test for significant correlations to the local temperature. – gavin]
John Fleck says
Re comment 2: It’s also worth pointing out that in the paper’s supplemental material, they show results of tests demonstrating that the results are robust when any combination of one, two or three of the series are arbitrarily excluded.
Re: current 3 [Fleck]
But John, these results won’t be robust until amateurs with Excel look at the raw data. On second thought, they won’t be robust to the amateurs after that either. Never mind.
Re #1 [PHE]:
Are there dendroclim scientists’ papers that you prefer that have been left out? Which ones do you like? Can you ‘splain why they should be included?
Also, which dendroclim scientists do you think are underrepresented in the literature?
Next, do you have raw data that we can audit that isn’t in the possession of the dendro scientists listed in the references?
On second thought: even a letter from spurned dendroclim scientists will do – is there a Googleable umbrage letter from the underrepresented dendroclim community, demanding their cores/corals/borehole/varve data be acknowledged and analyzed, and be part of the club?
[Response: We realise that this post and paper are treading on very well worn ground, but I would urge all commenters to keep the playground stuff somewhere else. -gavin]
Steve Latham says
“filtered to remove variations on time scales less than 20 years” in the caption — is this a 20 year moving average or weighted moving average? Is that why the analysis goes to 1995? Thanks.
[Response: From the Supporting Online Information:
The 14 proxy series were each smoothed to remove variations on time scales shorter than 20 years by the application of a Gaussian-weighted filter that reduces the amplitude of 20-year cycles by 50% and shorter cycles by more than this. Smoothed values were obtained up to both ends of each record by extending the records with the mean of the adjacent existing values.
Oh, also, is the sensitivity analysis (“robust with respect to the elimination of any single record”) just another way of saying they jackknifed the data? Would bootstrapping be better? I guess I’d better see if I can read the paper.
[Response: They’ve used the jacknife. Whether the bootstrap or jacknife is preferable in any particular situation is often a matter of taste, i.e. does one feel that resampling with or without replacement is more appropriate in addressing the issue at hand. In this case, I think the jacknife is the appropriate choice. Efron’s “The Boostrap, The Jacknife, and Other Resampling Plans” is a good reference for those who are interested. – mike]
Steve Latham says
Re: responses to 5 — thanks Mike, sorry I was so lazy. I’ve now read the supplementary info. Looks like their reason for jacknifing was that bootstrapping would have taken up too much time. Understandable. I generally prefer bootstrapping to jacknifing for assessing confidence in my own results, but seeing that their jacknife includes the exclusion of any three series (as noted in #3 above) is impressive.
[Response: Glad that helped to elucidate things. I think you’ve put your finger on why the jacknife was the appropriate choice here. Its a far more impressive result, and far more resistant to criticism, then if they had obtained the result through a bootstrap (where it would be unclear as to what precise pattern of exclusion of proxies contributed to the spread). And, depending on the number of surrogates used in a prospective bootstrap estimate, this choice was presumably less computationally intensive as well. – mike]
New question: what is the likelihood that new long term series will be developed to improve on this kind of work?
[Response: An excellent question. This approach may not be useful for quantitative reconstructions of past spatial patterns of climate fields, e.g. surface temperature, sea level pressure, drought, etc. However, it does allow hypotheses to be tested in a way that is free of the assumptions implicit in regression techniques. Indeed, the paleoclimate community, with help from the various funding agencies, e.g. NSF, NOAA, etc., is actively engaged in work that should extend our knowledge at the relevant temporal resolutions (i.e. decadal) several millennia back in time. The trick is finding proxy archives that can faithfully resolve the very low-frequency variability of interest on such longer timeframes. This is the challenge that the research community must take on. – mike]
Steve Bloom says
Mike, this seems like a good place to ask what you think of the recent Field et al study of fossil foramin populations in the California Current area (abstract). I notice that this is a quite different kind of proxy from the ones used in O+B. The authors seem to think that foramin trends in this area would tend to track global climate exceptionally well.
[Response:Yes, this is an interesting record. Quoting the abstract for our readers:
It is currently unclear whether observed pelagic ecosystem responses to ocean warming, such as a mid-1970s change in the eastern North Pacific, depart from typical ocean variability. We report variations in planktonic foraminifera from varved sediments off southern California spanning the past 1400 years. Increasing abundances of tropical/subtropical species throughout the 20th century reflect a warming trend superimposed on decadal-scale fluctuations. Decreasing abundances of temperate/subpolar species in the late 20th century indicate a deep, penetrative warming not observed in previous centuries. These results imply that 20th-century warming, apparently anthropogenic, has already affected lower trophic levels of the California Current.
The authors further add:
Our results indicate that the variability of foraminifera in the California Current in the 20th century is linked to variations in SST and is atypical of the preceding millennium. Given that the trend in global SSTs has been attributed to increases in greenhouse gases in the atmosphere (17–19), it follows that the best explanation for this ecosystem aberration is anthropogenic warming that has passed a threshold of natural variability.
So this record clearly adds to the weight of evidence that late 20th century changes are unusual in a long-term context. Of course, this is just one record, and a complex one to interpret. It presumably reflects changes in the California current and in regional upwelling. These features are connected, at least in part, with the El Nino/Southern Oscillation phenomenon, and ENSO has a somewhat counter-intuitive relationship with natural radiative forcings. Precisely how ENSO will respond to anthropogenic radiative forcing is still an area of active debate within the climate research community. – mike]
John Finn says
Regarding the reconstructions. There still seems to be a problem whereby the reconstructions are unable to emulate the thermometer readings. In fact, by the end of the 20th century, some are as much as 0.3 degrees cooler than the surface temperature record.
Which is incorrect – the reconstructions, the surface temperature record or both?
[Response:You appear to be misreading Figure 1 of our post which shows the instrumental series in black, and the various reconstructions as coloured curves. Some reconstructions do appear to underestimate the late 20th century instrumental warming (e.g. Moberg et al–the dark red curve), many others however track it well or even slightly overshoot it (e.g. Oerlemans–the maroon curve). So the basis for your assertion is unclear. And my response to your specific question would be “neither”, within the context of the estimated uncertainties in the respective curves. – mike]
I have not yet read this article, sorry, I will do so now.
I, being an anxious person, wonder if we have done a single tree ring data set from 800 to 2006 without using thermometers at all.
John Emery says
Nice editorial. Independent study?
I am interested in the strength of the relationship between tree rings and temperature. Do you know of any publicly available (i.e. URL) papers that go into this (I suspect this was the type of thing done a long time ago). Thanks.
[Response: Section 4.2 of the Jones and Mann (2004) review paper is a good place to start: Jones, P.D. and Mann, M.E., Climate Over Past Millennia, Reviews of Geophysics, 42, RG2002, doi: 10.1029/2003RG000143, 2004. – mike]
Arthur Reader says
Do you have information on why Osborn and Briffa used the controversial bristlecone pines when even the people who did the sampling (Graybill and Idso) made it clear that these measurements did not correlate very well with temperature?
It seems that whatever filtering Osborn and Briffa did make for proxy selection, it appeared to be rather flexible in terms of what got in and what was excluded.
[Response: Perhaps you need to read and/or re-read the paper and our description of the paper. Osborn and Briffa quite clearly describe an objective screening process which eliminated proxy series that did not correlate significantly with local instrumental temperature measurements over the 20th century. So what you mean by “rather flexible” is unclear. Their use of western U.S. tree-ring composites (which includes “Bristlecone Pine” data), as we understand it, specifically accounts for potential non-climate related effects originally proposed by Graybill and Idso, and explicitly estimated by Mann et al (1999). Corrected for these effects, the data correlate highly with local temperature measurements over the 20th century. They also use an independent, 2nd western U.S. tree-ring record that shows very similar behavior (compare panels 1 and 3 in their Figure 1) to the first. Furthermore, their conclusions, as described quite clearly in the paper, are robust to the exclusion of both of these, or in fact, any 3 of the 14 proxy series used. – mike]
Another question. I’m sure this has been debated at length so feel free to just point me to the appropriate blog/article. How do we know that tree ring widths are not just responding to increased CO2 and not increased temperature?
That paper “Climate Over Past Millennia” relies on the correlation of temperature with tree ring width as proved by other people/papers. I was more interested in these other papers that prove this correlation. I did see there was a reference to possible CO2 fertilization affect and it said this was compensated for, but with no detail of course since that is not the purpose of this paper (presumably the other people/papers have evaluated this relationship).
[Response: Section 4.2 of the Jones and Mann (2004) review paper was intended to be a starting place. If you read the various studies cited within that section, you’ll find lengthy discussions of these and many other issues related to the use of tree-ring data in climate reconstruction. I trust you’ll be able to find answers to all of your questions therein. – mike]
Rob Wilson says
We also published a similar paper this week:
D’Arrigo, R., Wilson, R. and Jacoby, G. On the long-term context for late 20th century warming. Journal of Geophysical Research, Vol. 111, D03103, doi:10.1029/2005JD006352
although our results are generally similar to the O+B and earlier studies, we cautiously conclude that there really is not enough data prior to ~1400 to make such definitive statements about comparing MWP and recent conditions – at least at these large scales.
[Response: Fair enough Rob. D’Arrigo et al is a nice contribution. But keep in mind that your conclusions were based entirely on a particular RCS tree-ring data set. Osborn and Briffa’s conclusions are based on an entirely different multiproxy dataset. So there is no proper “control” in this comparison. It might be interesting to submit your RCS tree-ring data to Osborn and Briffa’s methodology and see what conclusions are arrived at, no? – mike]
OT: What do you guys know about this allegation of Lysenkoism on hurricanes and GCC?
“Their use of western U.S. tree-ring composites (which includes “Bristlecone Pine” data), as we understand it, specifically accounts for potential non-climate related effects originally proposed by Graybill and Idso, and explicitly estimated by Mann et al (1999).”
How, specifically, was this estimate of non-climate related effects done?
My understanding is that the MBH99 estimate of non-climate related effects for the B_cone pines increased towards the end of the 19th century, then decreased in the 20th century. This would seem to be inconsistent with the known path of increasing CO2 concentrations in the atmosphere through both the 19th and 20th century, and the associated CO2-fertilization effect on vegetative growth.
[Response: Actually, not. It is well understood by those who study terrestrial ecosystem dynamics that there are multiple limiting factors on growth. For example, once a tree has essentially as much CO2 as it can use, other conditions such as soil nitrogen availability, will become limiting. In addition, the longer the stomates remain open (to try to take in the additional CO2), the more vulnerable the tree becomes to water loss through evapotranspiration. So one would only expect a significant impact of Co2 fertilization only until these other limiting factors kick in. Any subsequent increase in ambient CO2 concentrations would have little incremental value to the tree once this happens. A plateau in the observed response is the rule, not the exception. There is a vast scientific literature on this sort of stuff. We’ll leave it at that. – mike]
Armand MacMurray says
Re: response to 17
You note “There is a vast scientific literature on this sort of stuff.” Could you provide a reference or two to get us started? Thanks.
[Response: I trust you can find the information all by yourself. Try google scholar and start entering away. – mike]
Could climate change alter and/or accelerate the biological evolution of life? including human life. What kinds of selective characteristics would be favored in such circumstances? perhaps they would favorable, in that case anthropogenic climate change would be beneficial, would it not? After all green plants and plankton oxygenated the early atmosphere of Earth, creating an environment that allowed new life forms to emerge.
Mike – surely a bit of a mistake in your reply to 17 [nanny]. If CO2 levels are higher then plant stomata will tend to remain open for a *shorter* period of time since the plant will more quickly be able to absorb the CO2 required for growth.
In fact this effect [higher CO2 -> lower evapotranspiration] has been cited as a contributor to increased run-off in some river basins.
Your general point about there being multiple limits on plant growth is perfectly valid though. In fact the major cause of the positive feedback on the land carbon cycle in Cox et al was the stomatal response to water stress. [closing the stomata to prevent the plant drying out, thereby halting CO2 uptake and growth until water was available]
I also understand that, even in ‘ideal’ conditions where there is no nutrient or water shortage, there is a limit on the growth rate of plants such that plant growth responds less to CO2 increases above particular threshold levels [which vary from species to species].
On the other hand… it so happens that CO2 levels vary massively in the sub-canopy layer due to uptake by plants in conditions where air is not mixed downwards. Consequently a higher background level of CO2 will reduce the [albeit marginal] effect of these CO2 troughs.
[Response: Thanks, indeed you are correct. I got it backwards in my haste! – mike]
Mark #19 – Perhaps that would be true in an abstract ‘creative destruction’ kind of way. I would consider it very doubtful that the small mammals that scurried around the dinosaurs would have evolved into the vast panopoly of mammal species [including humans] were it not for the asteroid that caused the mass extinction at the end of the Cretaceous 65 million years ago. I’m less certain that the dinosaurs would have been willing to sacrifice themselves to enable the existence of cats and music by Beethoven [although I certainly like both of those things].
Life will certainly evolve. Gaia will respond. Another mass extinction event will throw open a multitude of evolutionary niches for species to exploit, undoubtedly producing an exciting array of new species. In, ooh, about ten million years or so.
My main point is that any major disruption of the Earth’s current biological systems will make it very difficult to support the existence of so many humans [that exist and feel pain, etc, right now] at the top of the food chain. In this context, serious, systemic change in the behaviour of the world’s climate and ecology is to be avoided at all cost.
Justin Rietz says
With all of these results graphed together, it is a little hard to read. However, several things I note:
1) Excluding the instrument data (which we can’t use to compare to 1000 years of data, anyway) it is apparent that we have seen temperatures comparable to today’s over the past 1000 years.
[Response:Untrue. Most of the reconstructions indicate late 20th century warmth that is unprecedented over the past 1000 years, without any consideration of the instrumental record. – mike]
Having the instrument data in black and covering the other trend lines, is, well, a little shady.
[Response:Watch the ad hom word choice. This is hardly “shady”. Most reconstructions have been calibrated against this target. It would be odd not to show how the predicted and observed values compare over the calibration period. Not showing that comparison might instead be considered innappropriate by some. – mike]
I think this graph would be more telling if this line was removed.
[Response:Most of these reconstructions are archived at the NOAA Paleoclimatology Data Center. There’s a link there from the RC front page. Anyone is free to download the series and plot them any way they like. – mike]
2) While it appears that the rapid increase in temperatures is unprecendented over the last 1000 years (see my #3), there is a corresponding drop in temperatures beginning around 1000AD. Wouldn’t it be reasonable to believe that the climate “pendulum” swung in one direction, and is now swinging back in the other direction as a response?
[Response:Not really. Much of the observed variability is captured by models forced with estimated natural (volcanic and solar) forcing in the past, and the added contribution of anthropogenic forcing in the 19th and 20th century. There is no evidence that the long-term radiative forcings follow an oscillatory pattern on multi-centennial timescales, though multidecadal (roughly 88 year) periodicities have been argued for in the case of solar forcing. The Jones and Mann (2004) review paper linked above provides a good discussion of all of this, and a good starting place for those interested in researching in greater depth. – mike]
3) The rapid increase in temperature apparent in the graphed data lines over the last 200 years or so is somewhat misleading: more recent data is more accurate, so it tends to correlate (and cluster) more closely, whereas the different data lines disperse the further back in time the chart goes, making temperature trends in the past look more flat.
[Response:There is no reason to believe that “more recent data is more accurate” with most of these reconstructions. Some use a time-dependent proxy data set, but others use a fixed set of proxies with reliable annual dating–no reason to believe the estimates are any more uncertain in the more distant past than in the less distant past in those cases. – mike]
I would be interested to see each of these temperature lines graphed separately – available some place?
[Response:See above. – mike]
J. Sperry says
Mike, I believe you mean Section 2.3 of your paper, not Section 4.2, in your responses above (#11 and #14). The question was about tree rings vs. temp., and Section 4.2 is largely about MWP and LIA.
[Response: I meant section 4.2, since this is where the issue of how tree-ring data have been used specifically in surface temperature reconstructions is discussed. However, section 2.3 provides some general background information on the use of tree-ring data in climate reconstruction more generally which might be useful to those looking to learn more about the subject. I would recommend this as a starting point, rather than the final word. The references contained within these sections provide a lot more detail on the subject. Thanks. – mike]
Re #18 (starting point for a “literature search” on plant growth response):
I recently took a course on ecosystem responses to climate change, and the recommended course literature included “Ecological Climatology” by Gordon Bonan (Cambridge University Press, 2002). I found this to give a pretty good (although somewhat wordy) overview of the title subject, and I learned a lot about how plants function and interact with their environment from reading the book. It might be a bit too detailed for what you want, Armand, but why not ask your library to find a copy?
Another good place to start looking for climate-related stuff is actually the good ol’ trusted IPCC 2001 TAR. The concepts of stomatal regulation (i.e. what Mike & Timothy were mentioning above) and how plant growth is influenced by climatic and abiotic factors is discussed in several places in the TAR, see eg. chapters 3 and 7 in the report of Working Group 1, as well as chapters 4 and 5 in WG2. (See the IPCC TAR index page, which also has a (limited) search function.)
“My main point is that any major disruption of the Earth’s current biological systems will make it very difficult to support the existence of so many humans [that exist and feel pain, etc, right now] at the top of the food chain. In this context, serious, systemic change in the behaviour of the world’s climate and ecology is to be avoided at all cost. ”
How do you propose that we prevent change?
Would someone please address Steve McIntyre’s assertion that none of these studies show late 20th century warming, unless they include a few controversial tree ring data sets?
[Response: One gets the impression sometimes that data sets are defined as controversial purely on their 20th century characteristics. The O&B study demonstrates that the issue of calibration is not key to determining whether the patterns of warmth seen in these long proxy records demonstrate an anomalous late 20th Century. Since the methodology (say Jones and Mann vs O&B etc.) isn’t so important for this conclusion, the argument then shifts to the suitability of the proxies. As has been said eleswhere, the number of suitable proxies for this kind of excercise is limited. The recent D’Arrigo paper has a few more that also demonstrate the same thing, and O&B’s result is robust to the removal of any three of the proxies. Since this methodology is relatively straighforward, any other proxies can be easily added in to the mix (and hopefully will be). The more data there is, the less important any single proxy will be – however, given the data we have so far, the result seems to be robust. -gavin]
Hank Roberts says
See everything in the sidebar, to start with. Read up on the subject of the site.
To “avoid systemic change” — don’t act in ways that may change how the system works. That’s not preventing change. That’s keeping it working.
Your circulatory system — you don’t want to prevent your blood from circulating, so you avoid systematic change in the way your heart functions.
As German rocket scientist Willy Ley said long ago, analysis is a fine tool, but you cannot figure out how a locomotive works by melting it down and analyzing the resulting mess.
Think about yourself — you no doubt “avoid systemic change” in your health — you don’t overeat, you don’t drink too much, you watch your cholesterol levels, you don’t start your car running while the garage door is closed.
Think about your car — you no doubt “avoid systemic change” all the time — you check the tire pressure, you use the right oil and change it regularly, you keep the level of brake fluid above the bottom of the reservoir so no air bubble gets in, you keep the cap on the brake reservoir so no moisture contaminates the fluid.
You work to avoid changing your system’s operation when you don’t understand how your system works, because you know it’s possible you may prevent it from working at all.
When it’s just you, hey, it’s just you. But if your family’s in the car, you probably drive a little more between the lines and keep the brakes in a little better shape, don’t roll a stop sign or take a chance on a railroad crossing. If you pilot an airliner, or operate a nuclear plant, “well, d’oh!”
Keep the needles out of the red zone, the hood ornament between the painted lines. Keep it moving, don’t screw it up.
Roger Raphael says
Please forgive the digression, but in today’s NYTimes, John Tierny claims that AGW can be mitigated by “geo engineering schemes to cool the planet by blocking sulight.” What’s he talking about? (On the Simpsons once Mr. Burns had a plot to blot out the sun, but it wasn’t a good thing.)
[Response: There is a 1992 NAS report: http://fermat.nap.edu/books/0309043867/html; see chapter 28 – William]
Re:28 I did see an idea whereby a large number of [large] ocean buoys would be manufactured that would float around and increase the amount of salt that would leave the ocean [not sure of the mechanism]. The idea was that the resulting increase in aerosol would increase the brightness and lifetime of low maritime clouds, increasing the planetary albedo. Seems to me that it would be more reliable to manufacture wind turbines and start to burn less coal and oil…
There’s also long been proposals to ‘seed’ the ocean with iron in order to increase removal of CO2 from the atmosphere by the ocean. I think they actually did some experiments with dumping iron and taking measurements of the consequent biological activity, but I don’t remember what the conclusions were. I don’t think it is going to add up to a “get out of jail free” card though.
Re #25. As #27 points out, the key word in my sentence is “systemic”. I propose that in order to reduce the risk of serious, systemic change , that we reduce GHG emissions. In time, we may well learn enough about the Earth’s climate to be able to purposefully make fine-tuning adjustments to our forcing in order to keep ourselves in a comfortable interglacial. At the moment, though, I think we should just stop poking the beast.
As a side note, there would be part of me that would be slightly disappointed were we to succesfully stabilise CO2 concentrations at ~400ppm. There’s a part of me that wouldn’t mind seeing how the experiment turns out. When would the Atlantic THC judder to a halt? How fast would the Greenland ice-sheet start to melt? I think we would learn quite a bit about various processes in the climate system by pumping out CO2 and observing what happens. Unfortunately this *isn’t* a lab experiment. We have to live here, so I think it is wise not to mess with it too much.
Stephen Berg says
Re: #26, “Would someone please address Steve McIntyre’s assertion that none of these studies show late 20th century warming, unless they include a few controversial tree ring data sets?”
The only way one would not see a late 20th century warming from these data sets is if you had the graph upside down. If McIntyre cannot see the late 20th century warming, I’d suggest he get his eyes checked!
Mark Shapiro says
Re: #25 and # 27 “How do you propose that we prevent change?”
We cannot prevent all change, of course, but my vote for the best take on energy policy is still with Amory Lovins at the Rocky Mountain Institute, http://www.rmi.org .
He’s been discussing energy efficiency and renewable sources since 1973, and looks at the economics, technology, and policies to move the technologies into the market.
Their latest book about transportation efficiency is on their website, available free in its entirety, as well as much other material.
As for Mr. Tierney in todays’ NYT, I would direct him to RMI as well. There are many reasons besides global warming risk to reduce fossil fuel use (more wealth, health, and security). Somehow Tierney thinks wind turbines and hybrid cars are expensive, but geo engineering will be cheap. Yikes.
Pat Neuman says
Instead trying to make sense of “controversial tree ring data sets”, I’ve spent much of my time looking at temperature data sets and fossil records from the late Paleocene and early Eocene.
Classic greenhouse warming with upper latitude amplification is shown in temperature 100-110 year plots for climate stations in the northern Great Plains, Upper Midwest and Alaska. Winter month average daily minimum temperature show the strongest warming trends. Summer month average minimum temperatures (July-August) are warming in central and northern Minnesota, the area where the moose populations have been decreasing at alarming rates … Kare11 – Rick’s special program on climate change in Minnesota, earlier this week, at: http://www.kare11.com/news/investigative/extras/extras_article.aspx?storyid=118105
Update – Plotted temperature averages (1888-2005: monthly, annual) for public view at:
Photos from Fossil Butte National Monument and pictures of late Paleocene and early Eocene can also be viewed at the link above.
Re #11 correlation of tree width with temperature.
For anyone that is interested I found a good paper that shows a very strong correlation here)
The correlation is amazingly good. It makes me want to go out for a trek here (British Columbia Canada) and try it out myself (with a tree stump that it has been cut down already by logging companies).
Georg Hoffmann says
Mike, excellent post!
I have a question about the representation of the tropics in those NH reconstruction. Are they not underrepresented in all Milleniums reconstructions? And is this not a general argument for a slight possible bias of all reconstructions towards overestimated variability.
[Response:Thanks Georg, a very good point. In a good number of the reconstructions, the sampling is largely of the extratropics, and the estimates largely represent extratropical temperature changes (e.g. Esper et al, 2003). In others, the sampling is in both the extratropics and tropics, but more weighted towards the extratropics (e.g. Jones et al, 1998; Mann and Jones, 2003; Oerlemans, 2005; Moberg et al, 2005). In yet others, there are more proxies in the extratropics, but an attempt is made to appropriately weight patterns of surface temperature that emphasize both regions, based on the relationships of those patterns with the available extratropical and tropical proxies (e.g. Mann et al, 1999). These differences in regional representativeness, rather than statistical methodology, appear to be the primary factor responsible for differences between various estimates (Rutherford et al, 2005). As alluded to in our post, one important issue is the possibility that changes in El Nino may have significantly offset opposite temperature variations in the extratropics, moderating the influence of the extratropical “Little Ice Age” and “Medieval Warm Period” on hemispheric or global mean temperatures (e.g. Cobb et al (2003). – mike]
On slightly different subject, could you guys give some insight into the cancellation of the Triana Deep Space Climate Observatory? According to an op-ed in the New York Times (subscription only), the cancellation may have been politically motovated and deprives the scientific community of an vital source of data on global warming. According to R. Pielke, Jr., it wasn’t really a high priority to begin with. What do you guys think?
According to Jeff Masters at the Weather Underground, the unusually strong ridge of high pressure that steered so many hurricanes to the US in 2004 and 2005 was formed in response to a strong warming of the ocean in the central North Pacific. He cites observations made in Dr. Bill Gray’s SUMMARY OF 2005 ATLANTIC TROPICAL CYCLONE ACTIVITY for this connection. Does anyone have any more information about this? (I know this is off topic, but the hurricane thread seems to have been terminated.)
Something has always puzzled me about the Wikipedia chart. The black line zooms high above all the colored lines at the end of the period.
1) Why don’t the colored lines go up too?
2) Since the colored lines don’t match the black line at the end of the period, what does it prove that the colored lines have historically been below the high point of the black line? To put it another way, what is the evidence that the method used to construct the colored lines is capable of detecting temperatures as high as the high point of the black line.
Thanks in advance for the educational lesson.
[Response: We didn’t create the Wikipedia plot, so you’d need to consult the authors of that page to get specific details. However, certain features are easily explained. In most cases, the reconstructions only goes through about 1980, because relatively few proxy series have been updated to the present (many of the proxy series were developed decades ago). So the interval of overlap between the proxy data and instrumental record is 1856-1980 or so. In many cases, the “skill” of the statistical model outside the calibration interval is established through statistical verification tests. For example, half of the (say 1856-1980) interval is used for calibration and the other half is used to test the performance of the reconstruction, and then the role of the two halves is switched. Such exercises also provide an estimate of the uncertainty in the reconstruction, which informs comparisons of recent instrumental changes with the longer-term reconstruction. The Jones and Mann (2004) reference discussed above provides detailed discussions of all of these issues, and links to many other studies. – mike]
Justin Rietz says
RE: Response #22
Thanks for the responses. One quick question: these plots are for northern hemisphere data only – isn’t this an inaccurate way to show “global” warming?
Mankind’s crime in global warming is that we time shift carbon oxidation, moving it forward, so to speak. All biology fundamentally sequesters carbon, providing fuel to drive the glacial cycle. We have too much carbon forcing precisely because the biomass was too energetic some 20,000 years ago.
Man, and other biological processes, take advantage of the carbon oxidation by controlling it and extracting work. The work, generally, is to sequester carbon.
Consider the old growth southern and northern forests. That carbon is not oxidized, it is sequestered in the great cities of the east and west coast, while the old growth has been replaced by carbon fixing grasslands and new growth. The new growth is sinking 2 million tons of carbon a year. So now we have basically four forests in sequester when nature might have just burned off the original two forests and triggered warming decades ago.
How much carbon have we locked up in irrigated soils?
Our crime is that we use carbon energy, and for each 10 atoms of carbon we sequester, one carbon escapes. We have so much carbon under management that the amount of energy needed to keep it sequestered is more than the atmosphere can contain.
What if man disappeared? Well, the glacial cycle would just oxidize the soils now left unirrigated, our great cities would rot releasing hindreds of tons of carbon, the new growth forests might burn off.
Robert Simmon says
I work at NASA Goddard Space Flight Center (not Goddard Institute for Space Studies) where Triana was built, but was not directly involved in the mission. In my opinion (and I’m speaking here as a private individual, not as a representative of NASA), the mission cancellation is neither surprising, or that much of a loss. Triana evolved out of Al Gore’s idea to replicate the famous Apollo 17 Blue Marble photograph, more for outreach than for science. The scientific community at Goddard (and likely NASA as a whole) was then asked to come up with instruments that would benefit for the satellite’s position at L1. I think it was regarded as potentially useful, but something of a diversion from the main sequence of NASA Earth observing satellites (such as the Landsat Data Continuity Mission, which is mandated by congress, but (inconceivably) not yet funded).
Since it was Al Gore’s idea, it’s been understood since November of 2000 that the mission was dead, even though the satellite was complete and sitting in storage. Subsequently, NASA Earth science as a whole has been battered, and things will almost certainly get worse before they get better.
Steve Sadlov says
RE: Bristlecone Pines. Being familiar with the area from doing both a bit of seismic reflection profiling there as well as recreation, I have to wonder whether Bristlecones at the tree line in the Whites and on the Eastern slope of the Sierra are more sensitive to moisture, or, temperature, in terms of their growth characteristics. What has made me wonder about this is firstly, the wide temperature swings known to occur in the area (trees living there are no doubt well adapted to dealing with a reasonably wide temperature range) and secondly the fact that 70 – 80% of the available moisture there is from snow pack. In other words, Bristlecones must also be well adapted to slurp up the moisture during a few weeks a year (like a cactus) and be more or less dormant otherwise. These characteristics raise interesting possibilities, for example, years of high growth may be years where there was a robust snow pack that lasted until late spring ranging into mid summer, years with less growth may have been warmer and drier. To wit, my seismic profiling schedule was once delayed by several feet of snow still there late May. That would have been a good year in terms of moisture availability during the growing season. Whereas, this year, being a La Nina year, pretty much sucks in terms of snow pack in the Whites.
Steve Sadlov says
RE: #33. Much of BC has a Marine West Coast climate. And at that, it is well north within that climate area. So, as a result, the frequency of drought stress is low, as compared with areas further south in that climate such as Oregon or Washington. More or less, the variable of moisture, at such a location, varies less, whereas temperature varies a bit more. Years that have multiple breakouts of the cold air pool that generally occurs in the Yukon-MacKenzie lowlands would probably cause the maximum stress whereas years where there was more onshore flow would be lower stress. So yes, I would expect decent correlation between temperture and growth characteristics in that locale. Plus one other thing. Trees that grow in Marine West Coast (and Mediterranean, for that matter) climates are not as well adapted to a wide range of temperature as would be ones in more continental climate types.
Hank Roberts says
> Triana … the satellite was complete and sitting in storage.
Steve Sadlov says
Setting aside for a moment any debate regarding proxies such as tree rings and formanifera, from the standpoint of cultural / archeological indicators (as well as historical accounts of things such as life ways, clothing, food and drink and the like) what are the things that tend to confirm and refute a global MWP of substantial degree (that term is somewhat loaded, but by “substantial” I mean things that have impacts similar to things thought to accompany the theorized coming warm up). For example, in Europe, people wore substantially thinner and less layered clothing during the MWP timeframe, versus what they wore during the Rennaissance, Enlightenment, Neo-Classical and Industrial Revolution time frames. That indicates the shift from the warm peak to the cold valley. Cathedrals were thought to be attractants to worship because they gave respite from the heat (even though some of them were not completed in time to reap the most benefit, we must look at the times of their initial conception / early project planning.) Take as another indicator the cultivation of citrus. Etc. What, if any, similar observations and anecdotal evidence are there in the Americas, Africa, Australia and Asia? I have not studied this and welcome what folks have to say in these regards.
Neil Fisher says
Re your response to #2: I know it’s not your work, but as a non-scientist, I’m curious about this aspect. If you need to “select” various trees that match local climate to use as a temperature proxy, then this obviously raises two points. 1) what percentage of trees show a good correlation to the instrumental record? and 2) what verification is there that trees that *do* match what records we have, have a similar correlation in the period of the reconstruction? It seems to me that without this information, it’s hard to judge what confidence we can place in the reconstruction. For example, if only 20% of trees show a correlation with known temperature records, it seems more like cherry picking the data than a reasonable proxy, but if the figure was, say, 97% it would be much more convincing.
[Response:These are not series selected at random from an unrestricted pool of potential predictors, in which case the issue of false positives needs to considered quite carefully. Rather, the proxies are a priori selected because they are believed to reflect climate parameters based on physical, chemical, or biological principles. The screening process, in this case then, is simply guarding against the use of proxies that one might have good reason a priori to believe should reflect local climate conditions, but in practice (e.g. owing to dating errors, within-sample reproducibility problems, etc) do not appear to show a strong signal. As for testing the reliability of the statistical model, that’s what verification/cross-validation is all about. See the Jones and Mann (2004) review paper cited above for a detailed discussion of these and related issues. – mike]
Mikel Marinelarena says
Sorry for lowering a bit the level of the discussion but 30 years into the most dramatic climate change that the Earth has experienced in the past millennium (perhaps since the beginning of the Holocene), I was wondering if this tremendous global warming should not have already become a bit more noticeable for the average person.
Right now, I know that most of the US and Canada are experiencing a mild winter but here in Europe we’re having one of the coldest ones people remember. Record-breaking temperatures in Germany, Ukraine, Russia,…hundreds of people killed in buildings collapsed under the weight of snow, hundreds (perhaps thousands) of people frozen to death in Central and Eastern Europe, the whole of Spain under snow blizzard alert,..
Apart from the few people living in certain Arctic regions, I’m not sure about anyone being able to claim that GW has become clearly noticeable for them in the past decades. The only thing most ordinary people would agree to is that the climate is no longer “what it used to be”. But then again I’ve been hearing that (and repeating it myself) since the sixties, when I was born. When have actually seasons been “what they used to be”? We have historical records of people complaining that the climate was going “mad” in 18th century Paris or even in ancient Rome.
It might just be anecdotes without any scientific value but the River Thames freezing regularly in winter, the Vikings being able to settle in Greenland in the Middle Ages or grapes being grown in Yorkshire sound like more plausible consequences of a climate change of the scale we’re said to be under. In spite of all the media bombardment about AGW I would say that the thing is going pretty smoothly so far. In fact, if one looks at the absolute maximum temperatures recorded for each continent, they all happen to have occurred before the last 3 decades: Europe->50.0 oC 1881, Africa-> 57.7 oC 1922, N America-> 56.7 oC 1913, S America-> 48.9 oC 1905, Antarctica-> 14.6 oC 1974, Asia-> 53.9 oC 1942, Australia-> 53.1 oC 1889. I can barely imagine the media hype if one of these events were to happen now!!
I guess that before the catastrophic consequences of the current GW appear in the coming decades, we can expect to start noticing the actual GW, can’t we? If so, when do the models predict that this might happen?
Of course we’ll have to wait some time to see an empiric validation of the IPCC estimates. But personally, I’ll find the trend of the following 5-10 years highly relevant to make up my mind. If the (slow) warming of about 0.18 oC/decade continues, the reliability of these predictions will strengthen somewhat in my mind, especially if I see the trend actually accelerating (as I understand most of the GCMs predict). But even in that scenario, I expect some skeptic scientists to plausibly argue that we’d just continue being inside a mostly natural trend, such as the 1910-1940 one or the end of the LIA. On the contrary, if I were to see a slowdown or even a reversal of the trend with an actual cooling of the SAT or even more significantly of the lower troposphere, I don’t see how I could reconcile that with anything I’m reading now by GHGs-enhanced AGW theorists.
I always welcome respectful corrections of my reasoning. Thanks.
[Response: Good places to start would be our glossary items on the LIA and ‘Medieval Warm Period’ (MWP), and the review paper by Jones and Mann (2004) cited above. -mike]
Robert Fawcett says
In response to Comment 46, 53.1 degrees C (Cloncurry, Queensland, 16th January 1889) is often given as Australia’s highest daily maximum temperature. Australia’s National Climate Centre does not regard this as being a credible observation for several reasons. Firstly, it is now known that the thermometer was not in a standard Stevenson screen at the time, but was instead placed inside a beer crate, pending the arrival of the Stevenson screen which turned up some weeks later. Secondly, all temperatures above 48 degrees C since 1957 have occurred in very different parts of Australia, indicating that the observation is climatologically anomalous. Thirdly, the observation is statistically anomalous when compared against nearby observations. Fourthly, Cloncurry is far enough north to hint that direct exposure of the thermometer to the sun may have contributed to the reading. Transcription error does not appear to have been a relevant issue. [See Trewin (1997), Australian Meteorological Magazine, 46, p251-256 for further details.]
The Climatic Atlas of Australia – Temperature, published by the Australian Bureau of Meteorology in 2005, gives 50.7 degrees C (Oodnadatta, South Australia, 2nd January 1960) as the highest daily maximum temperature observed according to current standards, with 50.5 degrees C at Mardie (19th February 1998) as the highest for Western Australia. The highest for Queensland is slightly lower, 49.5 degrees C at Birdsville (24th December 1972).
Some of the other continental extremes have also been disputed – see Krause and Flood (1997), ‘Weather and Climate Extremes’ (U.S. Army Corps of Engineers), for example.
[Response: Thanks for the info. The idea of doing met obs from within a beer crate seems very Australian :-) – William]
Steve Sadlov says
RE: #46. Indeed, so far as I can tell, the mean temperature in my part of California has cooled since I was a child. Also, it is now wetter here. We used to have really wonderful summers, quite warm but not opressively hot overall, puntauated by outbreaks of true interior type heat. Now, the summers are coolish and the outbreaks fewer. Winters were generally drier and cold was limited to a few outbreaks. Spring was warm to hot, with little to no rain after early April. Now, Spring has just about disappeared, the closest thing to it now would be the few warm spells happening during what is now, essentially, a winter, meaning both the chill and the rains, that goes from the first rains in late Fall until May, sometimes Memorial Day. We now crave and welcome those few “Pinnapple Express” set ups because although there is lots of rain, at least things warm up a bit. But generally, we only get those between December and February, with a more North to South Jet Stream set up after that. Another thing is that Fall happens earlier.
Today, we have snow (not sticking, thankfully) mixed with rain, down quite low, and during certain showers, down to sea level. The snow line proper is well below 2000 feet. Well, as a skier, I won’t complain …. looks like yet another year skiing into May in the Sierra.
Your ability to perceive temperature appears to be inconsistent with the data [fig 3 – irrigation has likely raised the GCV min temps and slowed the growth of the GCV max temps]. There is also a discussion of the UHI magnitude on the USGHCN in the paper, which should pique your interest, as well as this one which should help to improve your understanding of the UHI phenomenon.
Steve Sadlov says
RE: #49. Do *you* live on the Northern California coast? I simply shared what we’ve experienced here. For example, in terms of El Ninos, the strongest one in my own life had to be the 1982 – 83 one, and mind you we are talking Northern Cal here, not So Cal. The next strongest (in terms of “typical” effects – high snow levels and high precip) had to be the 96 – 97 one. But interestingly, by late 98 we’d flipped way back into the cold – snow at our place December of that year (highly unusual). And my coments about the summers getting crappy and spring essentially disappearing at my locale stand, those two El Ninos notwithstanding. Oh and by the way, we are expecting even lower snow levels next week. Stay tuned.