Doubling down on doozy: Anthony Watts is now denying the Little Ice Age

Professor Michael Mann recently wrote a response to a denier (who'd accused him of fraud), and corrected him, saying that:

Mr. Sauer begins by promoting the falsehood that “temperatures exceeded what we have today at a time (the Medieval period) when today’s industrialization did not exist”. That is so shopworn a myth that it ranks among the top climate change denier talking points (see the response to this myth by the scientist-run website Skeptical Science). The scientific consensus today is that, while some regions of the globe were relatively warm during the Medieval era, the warmth was not nearly as widespread as today. The overall warmth of the globe and northern hemisphere today is substantially greater than during Medieval time. Mr. Sauer might also want to take note that the year 2015 is off to the warmest start ever, 2014 was the warmest full year on record, and took place during the warmest decade on record.

A couple of days ago I wrote about how Anthony Watts used a wonky chart of temperatures on the summit of the ice sheet in central Greenland, arguing that it "proved" that Professor Mann and all the world's paleoclimatologists were wrong and he, science denying conspiracy theorist was right.

In support of his wrong claim that it was globally hotter in medieval times than it is today, Anthony tried to claim:

• the temperature of the ice sheet in Central Greenland in 1855 was a good proxy for global surface temperature of 2015. 
• if it was widely cold more than 11,500 years ago (the Younger Dryas), then it must have got hotter globally 1,000 years ago.

You say "huh? where's the logic in that?" and I say "there's none". Well, Anthony's doubled down, shifting his focus to a mountainous region in Spain (archived here).

Wrong location, wrong time

Today Anthony's back, in a vain attempt to rescue his tattered reputation. This time he's moved from central Greenland to north east Spain. He's claiming that global surface temperatures can be derived from temperatures high up in the Pyrenees in north east Spain.

Not only that, but he's claiming that a temperature reconstruction dating back to 1300 AD, proves that the medieval climate anomaly, that occurred from 900 to 1300 AD (probably peaking from 950 to 1100 AD), was global.

Someone needs to sit down with Anthony and give him a lesson in both history and climate. Someone needs to let him know just when the medieval era was. And they need to explain to him when the warming, which happened in parts of Europe, took place.

They also need to explain to Anthony the implications of him arguing that the Little Ice Age wasn't all that cold after all. And how one cannot draw any conclusions about global temperatures in the Little Ice Age from a temperature reconstruction based on trees in the Pyrenees in Spain.

Casting spells or casting doubt?

Anthony got a mite upset that everyone outside of WUWT views him as a laughing stock, so he decided to double down. He wrote:

Earlier this week, in Facts against the Mann, we noted how some ice core data cast doubt on Michael E. Mann’s recent claim that:
“overall warmth of the globe and northern hemisphere today is substantially greater than during Medieval time”

Except of course it didn't cast doubt despite Anthony wanting it to. Deniers never doubted that global warming isn't happening, or that if it is then it's "not us", and that if it is "us", there's nothing we can do to stop it, and if there is something we can do to stop it, then we mustn't. That covers some WUWT deniers. Others reckon that not only is global warming not happening, but an ice age is coming.

Deniers don't have any doubt. They might not have any coherent, consistent notion about climate, but they have no doubt in their often contradictory claims. So Anthony's silly article did nothing to their doubt. They had none before or afterwards.

Rather than casting doubt, I think someone may have cast a spell over Anthony Watts. Practically every time he writes an article he looks more and more stupidly ignorant. He shows that he doesn't understand history (he doesn't know when the medieval era was); he can't read charts; and he doesn't know the difference between local and global.

700 years of summer time in the Spanish Pyrenees - and the Little Ice Age

In this latest article, Anthony posted some charts of a summer-time temperature reconstruction of the Spanish Pyrenees and said:

Jim McIntosh writes on the WUWT Facebook page:
[A] recent paper suggests tree ring proxies (which underpin Mann’s Hockey Stick Graph that largely kicked off the global warming scare) have underestimated past temperatures, which creates the misimpression that current warming is unique.

The "recent" paper is by Jan Esper and colleagues, and is mainly about summer time temperatures in the Pyrenees, based on an adjustment of tree ring chronologies. The title of the paper is:

Long-term summer temperature variations in the Pyrenees from detrended stable carbon isotopes

And in the abstract the authors say:

Here we use detrended δ13C measurements in Pinus uncinata tree-rings, from the Spanish Pyrenees, to reconstruct decadal variations in summer temperature back to the 13th century. 

Nowhere do the authors make any suggestion that their paper is a global reconstruction. Of course they didn't. It was a paper reconstructing temperatures of the Pyrenees in Spain.

Last I looked, just as the icy cold Greenland summit isn't the whole world, neither are the Pyrenees in Spain the whole world. But Anthony decides to draw attention to a comment from a denier, on the WUWT facebook page, about a paper discussing the merits and difficulties of using carbon isotopes to help determine past temperatures.

The paper has more detail: the authors did isotopic analysis of tree-rings from 31 trees. The main reconstruction goes back only to the 13th century. That, as you know, is around the start of the Little Ice Age and around 200 years after the end of the Medieval Climate Anomaly. From the paper (my paras and emphasis):

Increment cores from living trees and dry-dead wood were collected (Fig. 1) to develop a continuous, millennium-length Pinus uncinata composite chronology for the Central Pyrenees (Büntgen et al., 2008).

The sampling site is located near lake Gerber (42°38'N, 1°06'E) at the upper treeline (2400 m a.s.l.), where traditional tree-ring width (TRW) and MXD data have been shown to contain a summer temperature signal (Büntgen et al., 2010).

From a total of several hundred Pinus uncinata sampled during several field campaigns, 31 trees were used to develop a stable carbon isotope ratio chronology spanning the past 700+ years with a fairly stable temporal distribution, ranging from 5 trees in 1300 C.E. to 13 in 2000 C.E. (bottom panel of Fig. 2).   

It's a study to reconstruct local temperature using tree-rings, based on istopic analysis of 31 trees near the tree-line in the Spanish Pyrenees. It is a tree ring study, discussing the merits of using carbon isotopes to help refine the temperature signal in tree ring analysis. Here is the bottom panel of Figure 2 that was referred to, showing the number of trees used in the isotopic analysis, and their age (click to enlarge it):

Figure 2 | δ13C age-trend detection, correction, and chronology characteristics (d) Temporal coverage of individual trees throughout the past millennium. Dark grey indicates the number of missing innermost rings on core samples. 

You can see why the authors didn't focus on the Medieval period. There is only one tree that that covers the period from 1000 CE to around 1220 CE.


What it is NOT:

• it is not a global reconstruction
• it is not a year-round reconstruction, it focuses on June-July-August
• it is not a reconstruction of temperatures in medieval times (it starts in 1300)
• it is not a multi-proxy study - it is solely derived from tree rings in the Pyrenees.

What it is:

• it is a temperature reconstruction of temperatures of the Pyrenees in Spain
• it adjusts the "raw" proxy data, modifying it in line with the results of their isotope analysis (note how Anthony didn't complain about the adjustment of raw data!)
• it is mainly on decadal time scales, not monthly or annual
• it is mainly about summer-time temperatures, not annual
• it is showing the Little Ice Age wasn't quite so icy in the Pyrenees in Spain as it may have been in other parts of the world.

Anthony Watts denies the Little Ice Age

Now deniers like to boost up the temperatures of the Medieval Climate Anomaly and drop down the temperatures of the Little Ice Age. In his WUWT article, Anthony does almost the opposite. He writes about a paper that doesn't start a reconstruction until well after the any medieval warming. And Anthony is trying to claim that it has "underestimated past temperatures". In other words, he is claiming that the Little Ice Age wasn't as cold as most reconstructions show it to have been.

Anthony wrote:

From the Esper et al. 2015 paper, note that the black line represents the instrumental record. The reconstruction, in red, shows Medieval Warming Period is clearly seen to have temperatures greater than the current modern period. As for the divergence of the instrumental record against the reconstruction, it is important to note that the trees sensing temperature aren’t near human habitation as most of our observing thermometers are, but rather in the Pyrenees mountains.

Here is the chart he showed, although I've copied it from the paper not from WUWT, and I've added some annotations. In particular, I've marked the peak period of the medieval climate anomaly, which goes off the chart to the left. I've shown the hottest time in the past at that location, being from 1375 to 1425. I've extended line showing the highest June July August temperature of the region across the chart, for both the instrumental and tree ring estimates. The chart shows that indeed the Little Ice Age doesn't show up at that location until quite late (the red line):

Fig. 5. δ13C based JJA temperature reconstruction and comparison with existing estimates. (a) Decadally resolved temperature reconstruction back to the 13th century derived from regression against JJA temperatures. Dashed curve indicates the period prior to 1260 C.E. when sample replication falls below three trees. The SE uncertainty band is derived from calibration trials against JJA temperatures using less-replicated chronologies to assess climate signal strengths during earlier chronology periods (see Methods). (b) MXD based JJA maximum (Büntgen et al., 2008) and May- September mean temperature (Dorado Liñán et al., 2012) reconstructions, plotted together with the regional JJA temperatures and uncertainty band from (a). All timeseries expressed as anomalies with respect to 1961–90. (c) Recent δ13C chronology tails for the raw data and data after application of the Atm, Kur, and Fen corrections to account for atmospheric 13C/12C ratio and tree physiological effects. 

In his last article on the subject, Anthony seemed to confuse the Younger Dryas, which ended 11,500 years ago with the Medieval Climate Anomaly. This time I don't know when he thinks the Medieval Climate Anomaly occurred but it wasn't in medieval times. That's off to the right of the chart and isn't part of the main reconstruction.

He also seems to think that the line showing instrumental temperatures (in black above) are artificially high toward the end because of the UHI effect. The authors on the other hand, calibrated the tree ring data against instrumental data. So Anthony is both accepting and refuting the paper that he claims proves something about global medieval temperatures.

It has been getting quite a bit hotter in summer in Western Europe

For an additional bit of trivia (not really trivia, but unrelated to global surface temperatures) here is Figure 4 from Esper15, showing how the authors calibrated the tree-ring data with summer data from the instrumental period of the twentieth century. As they point out, the decadal series has a stronger correlation but far fewer data points. The map at the bottom of Figure 4 also shows how summer temperatures in Western Europe have increased over the period 1901 to 2010. Victor Venema in the comments pointed out that the map is of the JJA temperature correlation field, not temperature increases. I should have read the caption I copied :(

Fig. 4. δ13C calibration against regional JJA temperatures. (a) Distribution and linear trends of the decadally (11 values) and annually resolved (109 values) δ13C and JJA temperature data over the 1901– 2009 period. (b) JJA temperature correlation field of the annually resolved δ13C data. Star indicates the location of the sampling site. 

Jan Esper and PAGES 2k

The article was authored by a team led by Jan Esper, head of the Department of Geography, Johannes Gutenberg University, Mainz (and Head of Dendro Sciences Division, WSL ETH Domain, Birmensdorf). He has done a lot of tree-ring based temperature reconstructions. Jan has contributed to the PAGES 2k project, co-authoring the 2013 paper in Nature Geoscience: "Continental-scale temperature variability during the past two millennia". I wrote about that paper at the time it was published. Deniers really don't like PAGES 2k work, because it shows that Michael Mann is correct when he says that globally it's hotter now than it was in medieval times.

Jan Esper and the Eurocentric Cartoon (Lamb's drawing)

Anthony might also take an exception the portrayal of the deniers' favourite IPCC drawing. Jan Esper was co-author of a paper calling the Lamb drawing in FAR a "Eurocentric Cartoon". You know the one I mean. It's from the very first IPCC assessment report in 1990:

In a 2010 paper about how surface temperature reconstructions have evolved and improved over time, Jan Esper and his coauthors wrote about the above diagram (my emphasis):

IPCC 1990: ‘Eurocentric Cartoon’
In the first IPCC report, presentation of temperatures over the past millennium appears to be a hand-drawn  sketch of how climate varied. A prominent MWP, followed by a cold period—the LIA—are observed with clear indications for high natural variability of the climate system. It was demonstrated that this figure was derived from the groundbreaking work of H. H. Lamb representing temperature change over Central England, but the chosen smoothing and figure description suggested more of an illustration, rather than quantification, of past temperatures. This curve presenting a Eurocentric schematic view of climate change, left room open for a completely new temperature history.

I don't expect Anthony Watts will be promoting that paper any time soon.

Global temperature reconstruction from PAGES 2k

To round off, below is Figure 4 a and b from the 2013 PAGES 2k paper, I've added some annotations (as an animation) so you can more easily see the 30 year average temperature up to 2000, as well as the periods spanning the medieval climate anomaly (900 CE to 1300 CE) and the little ice age (16th to mid 19th century in Europe). Click to enlarge it.

Figure 4: Composite temperature reconstructions with climate forcings and previous hemisphere-scale reconstructions. a, Previously published Northern Hemisphere 30-year-mean temperature reconstructions relative to the 1961–1990 reference period5, 43, 44, 45. b, Standardized 30-year-mean temperatures averaged across all seven continental-scale regions. Blue symbols are area-weighted averages using domain areas listed in Table 1, and bars show twenty-fifth and seventy-fifth unweighted percentiles to illustrate the variability among regions; open black boxes are unweighted medians. The red line is the 30-year-average annual global temperature from the HadCRUT4 (ref. 29) instrumental time series relative to 1961–1990, and scaled visually to match the standardized values over the instrumental period. Source: PAGES 2K (2013)

As you can see, with more and more studies being added, the picture of past global surface temperatures is becoming more clear.

Note however, that the reconstructions in the top panel are for the northern hemisphere only. The bottom panel is "across all seven continental-scale regions", so it's effectively global land surface temperatures (not including the sea surface).

Apart from some weak protests at the time the PAGES 2K paper came out, I've not seen conspiracy blogs refer to it. Deniers want to pretend that you can tell global surface temperatures from a single location, using tree rings or ice sheets, depending on their mood. (Usually they denounce tree rings, but they'll use them from time to time.)

From the WUWT comments

Early on someone pointed out that the reconstruction is from a single location. Luke wrote:

September 25, 2015 at 11:17 am
This reconstruction is based on one location so it does not represent a fundamental challenge to Mann’s global reconstruction.

Anthony Watts replied with a convoluted excuse, writing

September 25, 2015 at 11:23 am
Mann’s MBH 98 was heavily weighted on one proxy at one location, Sheep Mountain. McIntyre notes:
“Because Graybill’s bristlecone chronologies were so heavily weighted in the Mann reconstruction, demonstrating out-of-sample validity at Sheep Mountain and other key Graybill sites is essential to validating the Mann reconstruction out of sample.…In the Figure below, I’ve started with the Sheep Mountain chronology as used in Mann et al 1998 (left panel). As CA readers are aware, it has a very dramatic HS-shape and is heavily weighted in the MBH reconstruction. Indeed, without the Graybill bristlecones, the MBH reconstruction is basically noise.

Source [redacted by Sou]

So, “Luke” if Mann heavily weights a proxy in one location to make a global claim, does that make it OK for him, but not for anyone else?

There is so much wrong with Anthony's excuse that it's hard to know where to begin.

Let's start with Anthony deriding MBH98 for relying heavily on one proxy series while at the same time holding it up as an example that he should follow.

Next, MBH98 used a lot more proxy series. It doesn't just use one:

Figure 1 Data used in this study. a, Distribution of annual-resolution proxy indicators used in this study. Dendroclimatic reconstructions are indicated by ‘tree’symbols, ice core/ice melt proxies by ‘star’symbols and coral records by ‘C’ symbols. Long historical records and instrumental ‘grid-points’ series are shown by squares (temperature) or diamonds (precipitation). Groups of ‘þ’ symbols indicate principal components of dense tree-ring sub-networks, with the number of such symbols indicating the number of retained principal components. Sites are shown dating back to at least 1820 (red),1800 (blue-green),1750 (green),1600 (blue) and 1400 (black). Certain sites (for example, the Quelccaya ice core) consist of multiple proxy indicators (for example, multiple cores, and both d18O isotope and accumulation measurements). Source: MBH98

Another major problem is that Anthony knows that over the past 17 years there have been many more refinements to global reconstructions, by Michael Mann and by lots of other people as well. To use Michael Mann's very first reconstruction paper as an excuse is inexcusable. Since Anthony doesn't "believe" MBH98 anyway, and says so in his comment, why would he do exactly what he is criticising that paper for? It's worse than that. Although MBH98 did find that a tree series of proxies from North America were critical, that was by no means the only proxy series they used in that first paper.

MBH 98 was a very early attempt at a global reconstruction. The authors made it clear that the reconstruction was mostly based on proxy series from the northern hemisphere, and estimates of global temperatures were based on those plus a small number of proxy series from the southern hemisphere. For example, the actual "hockey stick" diagram was of the northern hemisphere only:

Figure 5 Time reconstructions (solid lines) along with raw data (dashed lines). b, for Northern Hemisphere mean temperature (NH) in 8C. In both cases, the zero line corresponds to the 1902–80 calibration mean of the quantity. For b raw data are shown up to 1995 and positive and negative 2σ uncertainty limits are shown by the light dotted lines surrounding the solid reconstruction, calculated as described in the Methods section. Source: MBH98

Next - just look at the 2σ uncertainty limits. Shortly afterwards, the same authors focus on just the Northern Hemisphere, in a paper with a title you'd never see on a denier blog: "Northern Hemisphere Temperatures During the Past Millennium: Inferences, Uncertainties, and Limitations."

You see, one thing you'll notice about climate conspiracy deniers, like Anthony Watts, is that they are so certain of everything. Scientists on the other hand will quantify uncertainties and describe the limitations to their work. Not so deniers.

The cavalier manner in which Anthony both derided MBH98 and decided it did the right thing after all is an example typical of conspiracy theorists, who are able to hold two contradictory notions in their head at the same time.


Rob Ricket is wrong when he wrote:

September 25, 2015 at 2:26 pm
Esper has long been a thorn in Mann’s side. Mann has on occasion attempted to bully him, but Esper has remained independent and resolute.

They may not always see eye to eye. However, when Michael Mann pointed out that a 2002 paper Jan Esper co-authored in Science mag was being misrepresented in various places, Esper and his colleagues agreed and wrote a follow-up paper, which clarified their work. And there was a beat-up by fake sceptics over a 2012 Esper paper, too. Michael Mann wrote an article about it at realclimate.org, and welcomed the work, saying it "represents a valuable contribution".


Michael Jankowski makes a wrong claim about another new paper, from Oceans2k, which I also wrote about recently.

September 25, 2015 at 11:27 am
The oceans2k results say we’re far from anomalously warm. Seems that Mann’s results contradict them…yet both are used to portray that man has has a large fingerprint in creating unnatural warming.

He's wrong because the paper isn't about current sea surface temperatures. It was mainly about the temperatures spanning the past 2,000 years. In particular, the cooling trend since 1200 AD. For the instrumental era, sea surface temperatures are shooting through the roof:

Data source: Hadley Centre, UK Met Office

Quite a hotch potch

Well, this article turned into something of a hotch potch, didn't it. Matching Anthony's flailing about all over the place. Not only did I benefit from learning more stuff from all the background research needed, I'm thinking it shows how much under the hammer is Anthony Watts. He isn't getting any support from his readers when he posts nonsense like this. And I'd say he's feeling the effect of being scoffed at by scientists, too (such as here and here).

References and further reading

Esper, Jan, Oliver Konter, Paul J. Krusic, Matthias Saurer, Steffen Holzkämper, and Ulf Büntgen. "Long-term summer temperature variations in the Pyrenees from detrended stable carbon isotopes." Geochronometria 42, no. 1 (2015). DOI: 10.1515/geochr-2015-0006 (open access)

Cook, Edward R., Jan Esper, and Rosanne D. D’Arrigo. "Extra-tropical Northern Hemisphere land temperature variability over the past 1000 years." Quaternary Science Reviews 23, no. 20 (2004): 2063-2074. doi:10.1016/j.quascirev.2004.08.013 (subs req'd)

M. Ahmed, K.J. Anchukaitis, A. Asrat, H.P. Borgaonkar, M. Braida, B.M. Buckley, U. Büntgen, B.M. Chase, D.A. Christie, E.R. Cook, M.A.J. Curran, H.F. Diaz, J. Esper, Z. Fan, N.P. Gaire, Q. Ge, J. Gergis, J.F. González-Rouco, H. Goosse, S.W. Grab, N. Graham, R. Graham, M. Grosjean, S.T. Hanhijärvi, D.S. Kaufman, T. Kiefer, K. Kimura, A.A. Korhola, P.J. Krusic, A. Lara, A. Lézine, F.C. Ljungqvist, A.M. Lorrey, J. Luterbacher, V. Masson-Delmotte, D. McCarroll, J.R. McConnell, N.P. McKay, M.S. Morales, A.D. Moy, R. Mulvaney, I.A. Mundo, T. Nakatsuka, D.J. Nash, R. Neukom, S.E. Nicholson, H. Oerter, J.G. Palmer, S.J. Phipps, M.R. Prieto, A. Rivera, M. Sano, M. Severi, T.M. Shanahan, X. Shao, F. Shi, M. Sigl, J.E. Smerdon, O.N. Solomina, E.J. Steig, B. Stenni, M. Thamban, V. Trouet, C.S. Turney, M. Umer, T. van Ommen, D. Verschuren, A.E. Viau, R. Villalba, B.M. Vinther, L. von Gunten, S. Wagner, E.R. Wahl, H. Wanner, J.P. Werner, J.W. White, K. Yasue, and E. Zorita, "Continental-scale temperature variability during the past two millennia", Nature Geoscience, vol. 6, pp. 339-346, 2013. doi:10.1038/ngeo1797 (sneaking realclimate.org access to the full article)

Most Comprehensive Paleoclimate Reconstruction Confirms Hockey Stick - article by Stefan Rahmstorf at Climate Progress.

Frank, David, Jan Esper, Eduardo Zorita, and Rob Wilson. "A noodle, hockey stick, and spaghetti plate: a perspective on high‐resolution paleoclimatology." Wiley Interdisciplinary Reviews: Climate Change 1, no. 4 (2010): 507-516. DOI: 10.1002/wcc.53  (pdf here)

Mann, M.E., Bradley, R.S., Hughes, M.K. "Global-Scale Temperature Patterns and Climate Forcing Over the Past Six Centuries", Nature, 392, 779-787, 1998.  doi:10.1038/33859 (pdf here)

Mann, Michael E. "Little ice age." Encyclopedia of global environmental change 1 (2002): 504-509. (pdf here)

Mann, Michael E. "Medieval climatic optimum." Encyclopedia of Global environmental change 1 (2002): 514-516. (pdf here)