Most Comprehensive Paleoclimate Reconstruction Confirms Hockey Stick

By Stefan Rahmstorf via Scilogs

The past 2000 years of climate change have now been reconstructed in more detail than ever before by the PAGES 2k project. The results reveal interesting regional differences between the different continents, but also important common trends. The global average of the new reconstruction looks like a twin of the original “hockey stick”, the first such reconstruction published fifteen years ago.

Green dots show the 30-year average of the new PAGES 2k reconstruction. The red curve shows the global mean temperature, according HadCRUT4 data from 1850 onwards. In blue is the original hockey stick of Mann, Bradley and Hughes (1999 ) with its uncertainty range (light blue). Graph by Klaus Bitterman.

78 researchers from 24 countries, together with many other colleagues, worked for seven years in the PAGES 2k project on the new climate reconstruction. “2k” stands for the last 2000 years, while PAGES stands for the Past Global Changes program launched in 1991. Recently, their new study was published in Nature Geoscience. It is based on 511 climate archives from around the world, from sediments, ice cores, tree rings, corals, stalagmites, pollen or historical documents and measurements (Fig. 1). All data are freely available .

Figure 1: The map gives an overview of the studied continental areas and the particular combination of the proxies used for each. Source: Nature Geoscience .

The climate history of the past one to two thousand years was reconstructed for seven continental regions in 30-year intervals (Figure 2).

Figure 2: Temperature evolution of the individual continental regions (30-year average). Red means hot, blue is cold. Source: Nature Geoscience.

Regional climate evolution

The data show, as expected, significant regional differences. Such regional patterns are an important clue to the causes and mechanisms of climate change. Significant local variations in climate can occur through changes in the atmospheric or oceanic circulation patterns, with very little effect on global mean temperature, because heat is only distributed differently (a recent short-term example is the record cold March in parts of Europe while Greenland was extremely warm – nothing unusual happened in the global mean). Changes in global mean temperature, however, occur due to changes in radiative forcing (eg solar fluctuations). This forcing can be globally uniform (e.g. well-mixed greenhouse gases) or can have a regional pattern (for example, volcanic eruptions and orbital cycles). So in the individual continents one expects a response to the forcings, superimposed by internal fluctuations.

The data shown in Figure 2 reflect this: they show some coherent signals, especially a long-term cooling trend that leads to increasingly cooler climate conditions from the relatively warm Middle Ages until this is turned around in the late 19th Century (see the next section). But, as expected, some regional variability is superimposed, especially on shorter time scales of decades to a century, where particularly warm and cold phases do not coincide in their timing on different continents, as the authors emphasize in the abstract:

There were no globally synchronous multi-decadal hot or cold intervals that define a worldwide Medieval Warm Period or Little Ice Age.

But they identify some shorter intervals where extremely cold conditions coincide with major volcanic eruptions and / or solar minima (as already known from previous studies).

Global Trends

The global mean temperature is of particular interest because it is a direct response to the global radiative forcing, buffered by the thermal inertia of the oceans. This follows from the first law of thermodynamics, i.e. the law of conservation of energy. So the globally coherent signals indicate globally effective forcings. The authors write in the abstract:

The most coherent feature in nearly all of the regional temperature reconstructions is a long-term cooling trend, which ended late in the nineteenth century. (…) Recent warming reversed the long-term cooling; during the period ad 1971–2000, the area-weighted average reconstructed temperature was higher than any other time in nearly 1,400 years.

The following Figure 3 therefore compares the area-weighted mean over the continents (b) with some previous northern hemisphere reconstructions (a) and the forcings (f, g, h). The basic pattern – a long-term slow cooling which in the late 19th Century turns into a rapid warming – has been known for 15 years and is often compared to a hockey stick: the long cooling trend is the handle, the modern warming the angled blade.

Figure 3: Temperature evolution according to some previous studies (a) and from the new PAGES 2k reconstruction (b). The panels f, g and h show the radiative forcing, see text. Source: Nature Geoscience . (Click here for enlarged image.)

Comparison with the forcings shows that this blade is due to the radiative forcing from the increasing amount of greenhouse gases in the atmosphere (green line in g). The solar and volcanic forcing can be reconstructed only with some uncertainty, therefore two variants are shown. For solar forcing, it is less the shape of the curve but mainly the amplitude which is controversial (a constant conversion factor) – the very high amplitude  adopted by Shapiro (dotted line) is widely regarded as highly questionable (see, eg, Feulner and Judge et al ). But even with this extreme assumption, the solar forcing in the 20th Century cannot compete by far with the greenhouse gases, and it also does not match the temperature evolution.

Interesting especially on long time scales of thousands of years is the orbital forcing (the well-known  Milanković  cycles; here is an online calculator for the radiative forcing). In the north, especially the summer insolation is important (green curve in panel h), because it is greatly amplified by the albedo feedback (ie, by changes in the ice and snow-covered area). This may explain part of the cooling trend in the Northern Hemisphere. In Antarctica the situation is different: in summer the ice-covered area is relatively constant at about the size of the Antarctic continent, because there is little sea ice, and the trend of solar radiation (blue curve in panel h) is relatively weak. Therefore the summer insolation change is much less important. But the annual average insolation in Antarctica (not shown in the PAGES graph) has declined strongly due to the orbital cycles over the past two millennia. This could be the explanation for the long-term cooling in Antarctica, which (unlike in the global average) has not yet been made up for by the modern anthropogenic warming.

Climate models are computer programs that calculate climate evolution from such forcings, based on the equations of thermo-and hydrodynamics. In the last IPCC report 18 such model calculations for the last millennium by different research groups are listed, almost all of which reproduce the reconstructed climate history from proxy data reasonably well (see IPCC, Figs. 6.13 and 6.14 – the red-dashed outlier in 6.13 is the drift-affected model used by von Storch et al. in their failed attempt to discredit Mann et al.). The models also show that without the anthropogenic forcing there would have been no warming in the last 150 years.

The first comparable hockey stick curve was published in 1998 and 1999 by Mann et al. – at that time based on data only from the northern hemisphere. Figure 4 compares the new hockey stick (from panel b in Figure 3) with this original hockey stick. (The data for the original hockey stick have also been available since 1999 on the NOAA Paleoclimatology website.)

Figure 4 Green dots show the 30-year average (area-weighted mean over the continents) of the new PAGES 2k reconstruction, as shown in Figure 3b. The red curve shows the global mean temperature, according HadCRUT4 data from 1850 onwards (also in Figure 3b, smoothed with a 30-year window). In blue is the original hockey stick of man, Bradley and Hughes (1999 ) with its uncertainty range (light blue). Graph by Klaus Bitterman.

For the scientific community, the confirmation of the old hockey stick is no surprise (except perhaps for the closeness of the match); many other climate reconstructions with a similar time evolution have already appeared since. Mann et al. at the time cautiously assumed a wide margin of uncertainty (light blue) because of their limited data base and a possible underestimation of the variance by their method; later reconstructions run largely within this margin. The work of Mann and colleagues has gained the highest recognition. For example, Bradley was honoured in 2007 with the Oeschger Medal of the European Geosciences Union and Mann likewise in 2012, and both were (as well as Hughes) elected as fellows of the American Geophysical Union. Politically motivated attacks on their work were immense, however – both Bradley and Mann have published books about that experience:

Stefan Rahmstorf is Co-Chair of Earth System Analysis, Potsdam Institute for Climate Impact Research. This article was originally published in German and translated was provided by the author.

31 Responses to Most Comprehensive Paleoclimate Reconstruction Confirms Hockey Stick

  1. prokaryotes says:

    What a great opportunity for the media to communicate these findings in a better way…

  2. prokaryotes says:

    What this study shows is that our so called warming of 0.8C is on top of the natural climate change, which would otherwise create the next ice age.

  3. Jacob says:

    “What this study shows is that our so called warming of 0.8C is on top of the natural climate change, which would otherwise create the next ice age.”

    Indeed. The pattern/cycle going back at least 800k years suggests that we should be entering into the next ice age within a few millennia. We have put an end to that cycle and are on the way towards something totally different.

  4. The links to the paper and data go to google translate, and it’s a bit awkward. Better to just link to the original page and let readers use Google translate if they want.

  5. Rabid Doomsayer says:

    Dream on.

  6. In this particular hockey match, it’s Michael Mann 5, climate change deniers nil. Good to see yet another validation of Mann’s ground-breaking contribution.

  7. Theodore says:

    If we could instantly stop all new greenhouse gas production (magically, I suppose) and at the same time turn off all natural feedback mechanisms (with the big green switch on Mother Nature’s giant climate feedback controller) how long would it take for the global average temperature to come to an constant equilibrium temperature? At what temperature would it stop rising?

  8. Mury Salby is getting a public spanking for doing bad climate science, and then he whines about it:

  9. Colorado Bob says:

    In a warming world extreme precipitation events will occur more often :

    “This is likeliest the wettest moment in Toronto’s history,” David Phillips, senior meteorologist with Environment Canada, said on Monday evening while rain was still falling. “By the end, we may have a new all-time one day (rainfall) record.”

    More than 90 millimetres of rainfall was recorded at Pearson International Airport in a two-hour period, starting just after 4:30 p.m.

  10. Too long. Too high.

  11. This study is just one more nail in the coffin of denialism.

  12. prokaryotes says:

    Canada Toronto “Record Breaking” Flooding 123mm of Rain July 8th 2013

  13. Mulga Mumblebrain says:

    Whoopee! As Country Joe might have said.

  14. Mulga Mumblebrain says:

    Too bad.

  15. Superman1 says:

    There have been a number of studies published assuming zero fossil fuel combustion ‘today’, with no positive feedback mechanisms operable. The peak temperature ranges from about 1.5 C to 3.0 C (with a few higher outliers), depending on assumptions for ‘climate sensitivity’, ‘aerosol forcing’, and other parameters. The peak is reached in about three-four decades, after which the temperature enters into a gradual decline.

  16. Superman1 says:

    What these results tell us is that elimination of all un-necessary fossil fuel uses NOW is a necessary, but not sufficient, condition to avoid the cliff. We need parallel carbon recovery and geo-engineering efforts to attenuate the peak temperature, or else we may still go over the cliff. Recent CP articles imply the possibility of even higher climate sensitivities based on historical data, so these ancillary measures become even more important.

  17. Michael Berndtson says:

    The problem is temperature anomaly. I realize it’s necessary for presenting data over an extended period with data calculated from ice cores, medieval records, thermometers, satellites and then averaged over a huge complex changing surface area called planet earth. However, an anomaly is really a temperature difference (delta T) between a basis point weighted for the later part of the twentieth century and the annual average.

    So what really is an anomaly? It took me several search iterations to find NOAA’s explanation. That was unclear and poorly written.

    If it takes an entire FAQ webpage on NOAA’s site to explain temperature anomaly, how can ABC, NBC, CBS, PBS, NPR, FOX, etc explain it in 30 seconds?

  18. MightyDrunken says:

    The temperature anomaly is simply the difference in temperature from a baseline temperature. Another way of putting it is the change in temperature.

    You don’t need to know how it was calculated.

  19. Michael Berndtson says:

    Yes you do. You simply restated what I wrote and told me that I don’t need to know any of the details. That’s the problem. If I think you sound kind of like a tool, why wouldn’t anyone else?

    Now we’re stuck with a graph that is essentially the derivative of temperature as a function of time. And the biggest sin by climate scientists is to plot temp anomaly along with measured carbon dioxide. That even makes less sense. The proper graph would be T and CO2 anomalies w.r.t time.

    The real problem is that temp anomaly smooths out the temperature rise over the past 30 years plus/minus 5. That’s because the moving average is taken into account and heavily weighted for data post 1981. And of course gives climate change deniers a reason to poke holes in the temperature increase vs CO2 increase argument.

  20. wili says:

    Part 1) Actually, some of the innovations the original Draco proposed were quite reasonable–being sure that all laws were prominently displayed, and distinguishing between voluntary and involuntary homicide.

    But what you seem to be proposing is much milder than Draco’s more famous, harsher punishments–capital punishment for even (what we would consider) minor offenses.

  21. wili says:

    Part 2)
    Done thoughtfully and intentionally, a steep reduction in GDP could coincide with full employment and mostly increased health.

    Unemployment is almost entirely artificial. If we had what every other industrialized country has–real national health care coverage–we could chop up most jobs so more people would work part time and so spread employment nearly universally.

    Instead of giving trillions of dollars to robber barons of finance, get them out of the FIRE business completely, grant full credit amnesty.

    In return for this munificence from the gov, the requirements to:

    fly hardly ever
    drive minimally if at all
    bike, walk and public transport the rest
    eat mostly plants
    live in smaller houses or with more people…

    may go down a bit more easily.

    The other things to do with what seems to be endless amounts of gov $$ now going to the robber barons would include:

    super insulating all houses and buildings
    starting relocation of pop’s near low coasts
    make everywhere more walk- and bike-friendly
    some scale up of mass transit…

    and of course continuing to increase alternative sources of energy

    Beyond this, we have to engage in a vigorous discussion of what life is for–

    consuming so much that we destroy our childrens’ (and even our own) futures?

    ‘cultivating our gardens’ (thanks Voltaire)
    supplying enough food and clothing, but much more than enough
    being sure that that is distributed well
    making community
    making music, dance, love (with contraceptives, please! ‘-)…

    Again, Britain reduced domestic use of petrol by 95%, yes, 95%! during WWII, domestic meat and dairy consumption plummeted…and by and large people got much healthier! And they won the war (with a little help from their colonialist friends).

    This ‘war’ is much more fundamentally existential than that one. Time to tell the truth about how dire it is, and get back to the idea that some things, such as our children and the future of the planet, are worth making (comparatively minor) sacrifices for.

  22. wili says:

    Part 3)
    I would also say that ultimately people are happier if they feel their life has a purpose beyond just self-gratification.

    Those who insist that we should never ask anyone to ever sacrifice anything for a viable future for their children are really banking on perpetuating an ultimately deeply unsatisfying culture of self aggrandizement, a culture that, even if it were carbon neutral, would end up killing both the planet and the human soul.

    It’s long past time to turn this ship around, away from the deadly fantasies of endless growth, and self-maximization.

    Who’s in?

    (Sorry for the long posts.)

  23. SecularAnimist says:

    Superman1 wrote: “elimination of all un-necessary fossil fuel uses NOW is a necessary, but not sufficient, condition to avoid the cliff.”

    You have been repeating that bumper sticker slogan for months.

    Yet you steadfastly REFUSE, when asked repeatedly by multiple commenters, to state specifically what these “un-necessary” fossil fuel uses ARE, or to set forth any actual plan or program for eliminating them “NOW”, or to tell us who will be responsible for eliminating “ALL” such unspecified “un-necessary” uses, worldwide, “NOW”.

    Unless and until you do that, you are just making meaningless and pointless noise.

    While making a lot of meaningless and pointless noise is sufficient to get attention, it contributes nothing of value to the discussion.

  24. red admiral says:

    In short: we need more super men (and women).

  25. Dennis Tomlinson says:

    “The true revolutionary “moves through the peasantry as the Fish does through water.” –Mao
    Or, for a second chance at AB:
    “The only distinction that democracies reward is a high degree of conformity.”
    –Ambrose Bierce, “Epigrams of a Cynic”

  26. Mulga Mumblebrain says:

    Joanne Nova! Much hilarity. Back to Andrew Bolt with you. Talk about a rogue’s gallery – Baron Monckhausen, ‘The Australian’, the extreme Rightwing and denialist (naturally) ‘Sydney Institute’, Murdoch’s most down-market rag, ‘The Herald-Sun’. My, what company you keep.

  27. BobbyL says:

    Didn’t the original hockey stick graph have error bars? As I recall, in his book on the climate wars Michael Mann emphasized how important the error bars are.

  28. Joe Romm says:

    They are here.

  29. BobbyL says:

    Thanks. I see them.

  30. Mulga Mumblebrain says:

    Sorry, Web-I seem to have gone off half-cocked. Ouch! I saw those dreaded words ‘Joanna Nova’ and assumed that you sympathised with her incontinent incoherencies. Many apologies.

Stefan Rahmstorf is Co-Chair of Earth System Analysis, Potsdam Institute for Climate Impact Research.