Sorry deniers, hockey stick gets longer, stronger: Earth hotter now than in past 2,000 years

The “hockey stick” graph is a reconstruction of Northern Hemisphere temperatures over the past thousand years. It showed a sharp rise starting about a century ago. Global warming deniers and doubters have long attacked the graph asserting that we were as warm if not warmer hundreds of years ago. But a 2006 National Academy of Sciences report largely reaffirmed the analysis.

A new peer-reviewed study by climatologists and earth scientists Michael Mann, Zhihua Zhang, Malcolm Hughes, Raymond Bradley, Sonya Miller, Scott Rutherford, and Fenbiao Ni now extends the reconstruction back nearly 2000 years:


Here is link to the Proceedings of the National Academy of Sciences study, “Proxy-based reconstructions of hemispheric and global surface temperature variations over the past two millennia.” The Supplemental Material is here (warning, big PDF). I have also taken some of the PDF figures and turn them into JPEGs. For the first ever, I believe, the authors did a multi-proxy reconstruction of the Southern Hemisphere for the past 1500 years (see figure at end).

A key advance of the new work is that it derives historical temperature through multiple, overlapping proxy records, including “the growth patterns of trees and coral, the contents of ice cores and sediments, and temperature fluctuations in boreholes.” Proxies are used because modern scientific instruments were available for only a small and recent part of Earth’s climatic history.

“Ten years ago the estimates for earlier centuries were really primarily reliant on just one sort of information: tree ring measurements,” said Mann of Pennsylvania State University.

“To satisfy the critics, we now have enough other sources that we can achieve meaningful reconstructions back a thousand years without tree ring data, and we get more or less the same answer”–that global warming is not mainly due to natural variability.

Some 1,200 proxy records were used, mostly from the Northern Hemisphere. These figures show the spatial and temporal distribution of the nine different kinds of proxy records used:


What are the paper’s main conclusions?

Following the suggestions of a recent National Research Council report [NRC (National Research Council) (2006) Surface Temperature Reconstructions for the Last 2,000 Years (Natl Acad Press, Washington, DC).], we reconstruct surface temperature at hemispheric and global scale for much of the last 2,000 years using a greatly expanded set of proxy data for decadal-to-centennial climate changes, recently updated instrumental data, and complementary methods that have been thoroughly tested and validated with model simulation experiments. Our results extend previous conclusions that recent Northern Hemisphere surface temperature increases are likely anomalous in a long-term context. Recent warmth appears anomalous for at least the past 1,300 years whether or not tree-ring data are used. If tree-ring data are used, the conclusion can be extended to at least the past 1,700 years, but with additional strong caveats. The reconstructed amplitude of change over past centuries is greater than hitherto reported, with somewhat greater Medieval warmth in the Northern Hemisphere, albeit still not reaching recent levels.

And what does the Southern Hemisphere reconstruction look like? The figures below are from the supplemental material, comparing the Northern Hemisphere reconstruction (top) with the Southern Hemisphere reconstruction (middle) — and a total planetary reconstruction is also thrown in (bottom). The Southern Hemisphere does not appear to show much of a Medieval warm period, based on admittedly much less data.


[As a side note, these figure match the overall shape of the Must have PPT #1: The narrow temperature window that gave us modern human civilization, though it looks a bit to me like that figure is more the Northern Hemisphere than the whole Earth.]

The NAS’s National Research Council report in 2006 was essentially a vindication of the original hockey stick, as evidenced by the Nature article on the report, “Academy affirms hockey-stick graph” and by the BBC article on the new study:

… a 2006 report from the National Research Council (NRC), commissioned by the US Congress, broadly endorsed its conclusion that Northern Hemisphere temperatures in the late 20th Century were probably warmer than at any time in the previous 400 years, and perhaps at any time during the previous 1,000 years.

But the 2006 report recommended using more and different proxy records, which the new study has. The bottom line from Mann:

You can go back nearly 2,000 years and the conclusion still holds–the current warmth is anomalous. The burst of warming over the past one to two decades takes us out of the envelope of natural variability.

This updated result will no doubt be widely attacked by the deniers, much as the original result was. But given the greater extent and diversity of the evidence, coupled with the superior analytical methods applied, I suspect this will be a definitive work for quite some time.

The planet is warming anomalously and dangerously thanks primarily to human emissions of greenhouse gas emissions, the bulk of which are from burning fossil fuels. The time to act is yesterday.

UPDATE: RealClimate has a post on this paper, “Progess in reconstructing climate in recent millennia.”

Related Posts:


45 Responses to Sorry deniers, hockey stick gets longer, stronger: Earth hotter now than in past 2,000 years

  1. paulm says:

    Ellesmere Island ice shelf disappears …

    A century ago, Ellesmere Island was covered with one continuous, massive ice shelf up to 70 metres thick and extending 500 km along its coastlines. Since then, about 90 per cent of the former “Ellesmere Island Ice Sheet” has disappeared – much of it during a warming period in the 1930s and ’40s.

    Good bye cold world your leaving me today this year!

  2. Dano says:

    This updated result will no doubt be widely attacked by the deniers

    Who cares? Of course they will attack anything that shines light on their discredited ideology.

    What society is doing now is not rehashing old, tired, discredited arguments (hint: the denialist fringe is not ‘society’), but instead debating the direction to go in. Currently the ship of society is on a course between 125º and 135º (denialists want 235º), and the discussions today – and on this site – are whether 125 is better than 135.

    Nowhere in these directioning discussions are debates about the Hockey Stick. The debates are about the grid holding up, corn for ethanol, drilling, insulation. There is no room in these debates for denialist drivel.



  3. Hugh says:


    The chirruping from CA is audible from across the pond!!

    I particularly like the flicking proxy graphs… with their implied message that you can define 1,700a of global trends by eye! Obviously, it’s something the whole family can do!

  4. Dennis says:

    It does matter that the deniers will attack this. The media, in its supreme stupidity, will talk about the “debate.” Someone somewhere will pull statements such as “strong caveats” completely out of context when talking about this. And the likes of FOX, if they mention this study at all, will only cover the deniers, not the actual science.

    Everyone who is given an opportunity to comment publicly about this needs to start by reading it completely, be sure to understand it, and then when the denier shows up on the other side of the panel, start out with the question: did you read the entire report?

  5. Paul K says:


    “The debates are about the grid holding up, corn for ethanol, drilling, insulation”

    How right you are! Whether you are an alarmist, a denier or anywhere in between, the focus should be on deployment, infrastructure and efficiency.

  6. Dano says:


    it matters only if you treat it as a “debate”. Change the tone.

    Treat it as ‘the same ol’ same ol’. Or ‘why beat this dead horse again’. Or ‘asked and answered long ago. Why must we revisit this same discredited argument yet again?’.

    The well-emplaced noise machine is breaking down – you know, the noise machine that is tuned to drown out relevant facts. Look at what is happening today in science, economics, politics. Time to start choosing relevant talking points to be in front of the issue, instead of reacting to their framing.

    Look at my first comment: who cares what they will say? We’ve moved on. Not ‘gosh, we’re still debating this’. No we’re not. We’ve moved on.



  7. Dano says:

    Oh, whoops: and what Paul K said.


  8. rpauli says:

    Why is it we can never find a statistician when we need one?

    How is the 0 degree level set? To me it looks like the Y axis for Temperature Anomaly is high by .4 degrees C

    Does one set the zero level by the entire data set? Or just the latest? Or the last few hundred years?

    To my eye the zero line should be way lower.

  9. Joe says:

    Rpauli — I’m sure the answer to your question is in one of the links I have provided.

  10. Dano says:


    what information do you have that any of the authors are lacking in statistics knowledge?

    Or are you simply spreading FUD without checking on the credentials of the authors?

    That is: your FUD doesn’t work here. Try somewhere that has gullibles as readers.



  11. rpauli says:

    My point is that this data LOOKS way too understated.

    If I had a refrigerator with an anomaly of only .4 degrees on paper… but most of my goods would feel a temperature change of over 1 full degree C.

    I shall look in the links for my answer. My question will be from an observation that the zero line seems to be the mid point between the highest and lowest.

    Where someone could put the zero through an average of the data and it would be much lower.

    Visually to me this appears that the temperature anomaly has already gone over 1 full degree C.

  12. rpauli says:

    And I just learned that “Everything is baselined to the period 1961-1990”

  13. P. G. Dudda says:

    rpauli: Eyeballing that chart, it looks to me like they *are* zeroing it on the long-term average. The maximum variations are about .9 below average and about .6 above average. It also appears to me that the standard deviation (again, eyeballing, no math) ranges from about +.2 to -.4 – so the fact that current temps are high enough to bump up a curve that was slowly dropping is… revealing, and should immediately ring “statistical anomaly” bells and trigger checks for measures of just *how* anomalous that temperature spike is.

    The short answer: AMAZINGLY anomalous. The technical answer: I’m willing to bet p values for how truly anomalous that value is ranges in the .95 to .99 range.

    Again, no math, just eyeballing. But that much is what I can make educated guesses on just using my memory of my intro-stats college course.

  14. Dano says:

    rpauli, please also include in your answer why you implied there was no statistical process analysis in the paper, and why you implied that they would be stupid enough to not include an author with statistics credentials.

    Thank you in advance.



  15. David B. Benson says:

    And in British Columbia, now warmer than at any time in the past 7,000 years:

  16. John Hollenberg says:

    David, the link above is broken.

  17. Not_At_All says:

    Mann’s new study proves nothing of that because it compares apples to oranges. The part that goes up and up in this century in the graph are real measured temperatures (CRU), not proxy information. If you want to compare what the proxies say about the MWP with something today, you have to compare it with what the same proxies say about today. Because if it is not the same than direct temperature measurements say, then the same could have happened in the MWP: higher temperatures than can be deduced from the proxies, so not good enough proxies, posibly because of acting like a low-pass filter on temperatures, therefore showing a mild increase that would hide faster, say 15 year fluctuations up and down.

    If you look only at the lines ofthe graph that correspond to proxies, you will see nothing strange or unprecedented in modern times.

  18. jorleh says:

    Fine work, Joe. I wonder, if the energy experts McCain and Palin take some notice of this science.

    This was a joke, but perhaps there are some more energy experts who took it seriously. To be serious, don´t elect idiots to destroy us all.

  19. PHE says:

    “Earth hotter now than in past 2000 years”. Please explain how this is demonstrated. The proxy temperatures at the end of the graph are exceeded by earlier proxy tempertaures. The instrumetnal data of the ‘present day’ (or as near as it gets) is upt to 0.7 deg above the present day proxy. Therefore, the real temperature could have been ) at least 0.7 deg above or below the proxy line at any time in the past.
    If I’m wrong, please explain.

  20. Dano says:

    Please explain how this is demonstrated…If I’m wrong, please explain.

    The squiggly line thingie recently is mo’ closah toward the top of the chart doo-hickey on the right than the squiggly thingie toward the left. This is glaringly apparent unless you are a denialist – in which case, your brain is doing convulsions to avoid the information and you can’t see it.




  21. PHE says:

    Dano, of course its glaringly obvious that the line on the far right is closer to the top than the line on the left. What is not glaringly obvious to the casual observer is that this line on the right (the thick red one) is instrumental data (as is the pale grey one just behind it). Its incorrect to suggest you can compare this directly with proxy data. If you doubt this, then you need to explain why the proxy data and instrumental data during the most recent years (with the most detailed, reliable and accurate data) do not match.

  22. David B. Benson says:

    For those who have trouble with this thread’s graph, I previously posted a link which shows it is now hotter i8n British Columbia than at any time in the last 7000 years.

    Here is one for the Alps and 5500 years:

  23. Dano says:

    What is not glaringly obvious to the casual observer is that this line on the right (the thick red one) is instrumental data (as is the pale grey one just behind it).

    Ah. So the casual observer went to a school that didn’t teach its students how to read charts and maps.

    There is a thing called a key (the thingy occupying the upper 20% of some graphics, 40% of others) that describes the colors. I see they had some graphics assistance, as the colors are the same and have the same placement and arrangement throughout the three series to lessen confusion.

    Its incorrect to suggest you can compare this directly with proxy data.

    I look forward to your letter in PNAS explaining why this line is problematic. Let us know if it gets accepted.

    Otherwise, the instrument-proxy analysis can be found in the paper, pp 13253 and 255, in the Data and Results section. Else the wider paleoproxy literature explains this, specifically the portion of the literature that reconciles proxy data to instrumental data, and explains why the red line sits where it does wrt the proxy lines.



  24. PHE says:

    Dano. Well you clearly don’t have an answer. Its not a trick question. To accept that the historical proxies are a reliable representation of temperature on a year-to-year basis, then you’d expect them to match during the most recent period when reliability and number of measurements are greatest. I accept I must be missing something in the understanding, so if you can help explain it that would be useful. If you can’t, then I would have to question whether your acceptance of the answers is based on understanding or faith.

  25. Dano says:

    I told you where the explanations were, by page in the paper in question. Other readers can read, aren’t stupid and won’t be fooled by your hand-wave or fake argument from ignorance.

    If the temp record is inaccurate, send a letter to PNAS and show us how you are Galileo.

    Spare us your patently transparent tactic. Only the gullible dupes believe it, and they don’t get access, so why are you bothering?



  26. PHE says:

    Oh well. When personal attack overides argument, it doesn’t say much for your own self-belief.

  27. Dano says:

    Ah. Chronicling rhetorical tactics is personal attack. Sure. Take a rhetoric class, lad.

    I guess this is the best the denialists can do.



  28. PHE says:

    The Mann et al paper aknowledges the problem: “we observed evidence for a systematic bias in the underestimation of recent warming….the observed warming rises above the error bounds of the estimates during the 1980s decade, consistentwith the known ‘‘divergence problem’’ “. It concludes the problem is reduced through “elimination of all tree-ring data from the proxy dataset”. The best result is achieved with the EIV approach which uses “nonlocal and NON-TEMPERATURE RELATED proxy forrmation …thereby AVOIDING RELIANCE ON PURE TEMPERATURE PROXIES that may exhibit a low-biased sensitivity to recent temperature change.” [my capitals].

    This sounds very strange. We are asked to accept that the proxy data represent a reliable record of the past. Yet, the paper admits that it cannot accurately represent the present, and further, cannot explain why.

    This is a perfectly serious observation which I’m interested in seeing a serious response to. (Don’t bother replying Dano unless you have something grown up to say).

  29. Dano says:

    If you think your perfectly serious observation is so devastating, send in your letter to PNAS and share your perfectly serious observation that you know something that the rest of the world should be perfectly seriously observing.

    You have an obligation to society to tell us your perfectly serious observation. It is not spread to society in blog comment threads. It is spread to society via the literature. It is up to you to stop the madness that your perfectly serious observation will shed light on.

    Let us know if your letter is accepted so we can watch the journal for this event.



  30. P. G. Dudda says:

    Dano, I think that PHE is asking “what calibration has been performed on the recorded temperature data that allows the authors to use it on the same chart with the proxy data?” (I don’t know the answer, I’m just attempting to rephrase the question.)

  31. Dano says:

    PGD, I answered that already, giving the page numbers of the paper and the general literature for larger context wrt whether it is possible (there is even discussion in the paper regarding some of the issues with datasets and calibration.

    Perhaps you were confused by the false assertion that I didn’t answer the question & that led to your restatement.



  32. Mark Zimmerman says:

    This sounds very strange. We are asked to accept that the proxy data represent a reliable record of the past. Yet, the paper admits that it cannot accurately represent the present, and further, cannot explain why.

    It would have helped a lot if they had continued the EIV plot into the 20th century. I asked about this same issue on Realclimate and was told ‘it’s the plotting’ and was invited to plot the EIV data into the 20th century myself.

    I’m not a denialist; just had a denialist break my balls on this very question in another forum.

  33. PHE says:

    BTW, Dano gives me too much credit. When it comes to ‘personal attacks’, he does quite a good enough job himself of making his ‘arguments’ look ridiculous. I take no credit for that.

  34. Dano says:

    Personal attacks (eg ad hom): you are wrong because you are an idiot.

    NOT personal attacks (NOT ad hom): you are wrong because of A, B, C, and by the way, you are an idiot.

    Your rhetoric needs work. A lot of work. Write the authors and tell them they are wrong. Write PNAS and tell them too. Let us know how that turns out.


    Mark Z:

    I read the exchange you refer to and in the page numbers I give above and explain, the issues with datasets and calibration that I mention and gavin talks around is generally known as the _divergence problem_, discussed in the paper.

    Mann has effectively said – in my view – this is our analysis (second author is a stats guy), this is what we came up with, there are the data. It sounds like (I haven’t corresponded with any of the authors) other things are in the works and rehashing old ideas is being put to bed. As the paper has lots of links to data and the datasets are on line, if folks want to do more work, they can have it. Hence the tenor of my responses above.

    There are many more things to do that take importance over rehashing old ideas. Folks with natural science degrees (I am one multiple times) see many signs that cause great concern. Anyone visiting a decent Uni library and spending a day in the journal stacks sees the same thing.

    That said, adding more data to allow us to better map scenarios for adaptive management is a good thing. The world’s militaries are already doing this.

    But the point is: we have moved beyond this quibble. We are trying to figure out what to do in a warming world. Nailing down precisely how much man contributes to this (most of it) is helpful to craft good policy. But it is not helpful to get society going, which is what needs to be done now, so the details about abatement can be accepted. Without acceptance, nothing gets done. This is how it works. Folks quibbling about details of minor importance don’t want the societal discussion to happen at all. That is what we see here. And on the other thread, with that snowblind character, you see the effect of how the game is played to delay the discussion – it’s to the point where you don’t know if this is a gullible dupe or a misinformation campaign to create a false sense of a grass-roots movement – believe me, that is the main thing in the way of decision-making, not a question about the science.



  35. Mark Zimmerman says:

    As the paper has lots of links to data and the datasets are on line, if folks want to do more work, they can have it. Hence the tenor of my responses above

    I would do this if I had the ability; can you point me to where I might get the program to download and display the data Gavin directed me to?

  36. PHE says:

    As summarised above, Mann et al state they have no explanation of why the instrumental data are so much higher than the proxy data in the most recent times. Without this explanation, a logical conclusion from the proxy graph is that the MWP was at least as warm as now and very possibly warmer.

    Take the difference between instrumental and proxy for the most recent time: approx 0.7 deg C. Then add this to the highest proxy result in the MWP (0.2 degC). If there were relaible thermometers then, it is perfectly possible that they would have recorded up to 0.9 degC (on the given scale). Of course there is no proof, just a possibilty.

    It is therefore difficult for an idiot to see how Mann et al draw their conclusion “that recent Northern Hemisphere surface temperature increases are LIKELY anomalous in a long-term context.”

  37. kuen says:

    this is sience fiction!

    the hockeystik ist nonsens, completely wrong. everbody should know about temperatures during lia and the warmer periodes.
    what you show, is dramatically in pulling down the warm times and useless uprising in the lia temperatures.
    i do not talk to peoples any more, how do not understand simple climatologie at all. this is only dramatik fiction, no sience.



  38. Janama says:

    “Without this explanation, a logical conclusion from the proxy graph is that the MWP was at least as warm as now and very possibly warmer. ”

    you mean like in this chart from this same site that shows the MWP at a higher temp than today.

  39. shop says:

    Some 1,200 proxy records were used, mostly from the Northern Hemisphere. These figures show the spatial and temporal distribution of the nine different kinds of proxy records used

  40. Bob Herron says:

    The Mann hockey stick is totally discredited by scientists and statisticians around the world. Just as the last ten years is collapsing the global warming religion (now climate change) the science generated by warmists is being uncovered as sham and data manipulation. The latest being the claim Antarctica is warming, discredited even by global warming believers.

  41. Mary Hinge says:

    So Bob,
    I see you have been trawling the depths of the denialists’ blobs, the language and phraseology is exactly what we would expect.
    Perhaps you would like to share your references for the ‘hockey stock’ being ‘totally discredited’ by scientists and statiticians? (not blog references, I mean actual scientific references) As far as I am aware over 99% of credible scientists do not share this view and statiticians, well Steve McIntyre comes to mind….no…can’t think of anyone else.
    Bob, remember that all the credible scientific papers in the last 20 years show that the affects of AGW are increasing and the fact that man is the major cause is beyond reasonable doubt.
    To learn more about this subject broaden your reading. Virtually all of the denier blogs are run/supported by the religious right wing (your ‘global warming religion shows how they are trying to delfect attention from this) and oil and energy companies. If you don’t believe me about the religious right then just look at the denialists websites when Obama was in the process of being elected, very illuminating.
    I hope you will heed the advice, read more from peer reviewed journals and less from the same sources that deny evolution and’or are convinced of a huge inter-governmental cover up (very similar in style to those who believe in Roswell, tin-foil hats and grassy get the picture)

  42. KW says:

    So, as of 2009…aren’t we back down to the peak of the Midevil times? It would seem so with all the people wearing chain mail and such, jousting each other over trival things such as tenths of degrees of warming.

  43. GS says:

    70% of statistics are made up on the spot

  44. JC says:

    “Virtually all of the denier blogs are run/supported by the religious right wing (your ‘global warming religion shows how they are trying to delfect attention from this) and oil and energy companies.”

    I am quite happy to read both sides to make my own mind up, but this annoyed me. The leaked CRU emails clearly show that Jones etc were courting amongst others Esso (exxon), so lets drop the “holier than thou” stance.