Taylor Dome CO2 used by IPCC, corrected timeline

Posted: November 12, 2015 by tchannon in Analysis

During June 2014 oldbrew published “Explaining(?) abrupt climate change” based on an article published by Judith Curry. The author noticed Antarctica Taylor Dome is mentioned, an interest of the author because of long known highly suspect features of the data for this core and realised the implications of cross checking Antarctica ice cores to a common major event history, in this case with a profound change in timescale. Assuming the high resolution ion data and the low resolution gas data share time, reworking ion data also reworks gas data. The effect of doing this moves CO2 more recent and produces a good CO2 modulation match for the Medeaval Warm Period and Little Ice Age . The IPCC use Taylor Dome as part of the CO2 story. A lot of information has been omitted from the below such as the history over Taylor Dome dating conflicts. The original draft was written a year ago, now been shortened.

The paper behind the above is published Nature Climate Change, “Insights from Antarctica on volcanic forcing during the Common Era” doi:10.1038/nclimate2293. A supplementary file is available containing sufficient information for a reproduction of the critical part.


Figure T1, final result, reproduced with comment later.
Whether this is valid is a matter for discussion.



Figure T2, Reworked data showing the old and new timescales.


Ice cores are infamously a mess on dating regardless of claims by many to the contrary. This certainly applies to Antarctica. The paper is attempting to produce a consistent timeline between all the Antarctica cores, maybe 10 based on common volcanic event markers. This is done in hindsight with access to multiple datasets rather than an attempt to put a timeline on a new data.

One of the cores brought into line is Taylor Dome which is at an unusual location rather close to the ocean (see fig Tx).

The above is a large adjustment and means the ice is much newer than stated in the original publications.

I set about reproducing their result, at least near enough as a confidence check.

Figure T2 is the result, near enough for my purposes and included resampling to regular.


Figure T3

This is the scheme I used, crude but works. No attempt was made to handle curvature, which seems minimal over the period of interest.


Figure T4, annual data plot

There is another ion data concurrent with the Sulphur used for the volcanic eruption work. Figure T4 includes Na. Sodium. I was surprised to see a major peak very close to AD1540. This is a curiosity at a near ocean Antarctica site.



Any connection with a European event seems far fetched but if any readers have information on a global event this would be novel. It also presupposes remarkable timing accuracy which I rather doubt.

The matter of Taylor Dome carbon dioxide proxy as used by IPCC justification of human influence

Taylor Dome is also critical to IPCC assertions


Figure 10 from IPCC TAR WG1

Figure T5

The above remarkably vague plot is of a critical claim about history. The new work on an allied data shows the timeline in error. I think major timing error was at least suspected when the ice core original work was done, moreover other works shortly afterwards all point to error. This does not seem to have been addressed.

Also available as a PDF (2.5MB) and PDF are often as vector plots, indeed so, therefore…


Figure T6
Overlay plot showing the author has accurately identified the dataset used for the IPCC plot.


Figure T7, showing the substantial timing error between IPCC claim and the ion based retiming

Now I overlay this new result on a widely shown plot of estimated temperature.

From PennState archive

Medieval Climatic Optimum
Michael E Mann
Volume 1, The Earth system: physical and chemical dimensions of global environmental change,
pp 514–516
Edited by
Dr Michael C MacCracken and Dr John S Perry
Encyclopedia of Global Environmental Change
(ISBN 0-471-97796-9)
Ted Munn
(c) John Wiley & Sons, Ltd, Chichester, 2002



Figure T8

A timeline correction places the CO2 trace sensibly aligned with H. H. Lamb’s MWP and LIA, however, I believe little evidence has appeared showing either to be global or affecting Antarctica although recent works suggest there was an LIA effect in Antarctica. However, the absence of a temperature effect at Taylor Dome is unrelated to recording gas values with causal elsewhere, gas being fairly well mixed in the atmosphere.

Paul Homewood has in this December 9, 2014 article saved me the trouble of digging out references. See this blog article.

The Antarctica circumpolar currents tends to equalise coastal climate.


Figure T9 from http://cdiac.ornl.gov/trends/co2/ice_core_co2.html



Figure T9 Courtesy of Warwick University “Antarctic surface currents and surface wind patterns. The surface winds drive the currents (modified from Sugden, 1982).”

Turning to CO2 data for Taylor Dome

First I point out my long familiarity with Taylor Dome data, originally brought by seeing dubious features of the data, improbable.


Figure T10

Ice cores are dubious things especially where structures move/flow as is likely here.

Snow accumulation and compaction follows a power law but this seems to vary with overburden. There is supposed to be a gas pore closeoff point when gas interchange with the atmosphere ceases but this and the effect of gas forming other phases leads to very questionable data as a proxy. I don’t want to get into that argument here.


Figure T11
This law has long struck me as peculiar, a ridiculously close to square law large section of the core. This does tend though to confirm the pore shutoff depth claim of 70 metres. (fit r2 = 0.9995)


Figure T12, as figure T11 but for Ion data, note very extended deep core but also the commonality of shape but also a subtle difference in depth axis. Maybe my mistake.


Figure T13, very top of ion core is peculiarly linear given snow compaction, Sorge’s law. This may be helpful, Bader has many papers



Figure T14, all recent samples on large step, two methods shown, one trivial, the other suppresses noise, same result



Data ftp://ftp.ncdc.noaa.gov/pub/data/paleo/icecore/antarctica/taylor/

Supplement http://www.nature.com/nclimate/journal/v4/n8/full/nclimate2293.html?WT.ec_id=NCLIMATE-201408#supplementary-information (344kB)

Post by Tim

  1. Bloke down the pub says:

    So now they can point to the evidence from a nonexistent mwp that CO₂ increase causes warming?

  2. rishrac says:

    Bloke that was the idea, the MWP and LIA never happened, only locally not world wide. Recent drilling world wide reflects a very different story. Both events were world wide, with many smaller events. In any event the science is settled since CAGW just ignores errors and other conflicting data. co2 does everything bad and nothing good. If I’m not mistaken one of the graphs had co2 at 180 ppm, that’s plant death at those levels. Part of the rationale that the MWP never happened is that Antarctica didn’t melt and lead to much higher sea levels. Unlike today where at any minute the ice sheets are going to give way, wait, what NASA, what are you saying? A warming world and more ice? This is a conundrum.

  3. tallbloke says:

    Great post Tim, much to think about. First thought: How come the MWP only raised CO2 by a few ppm? Answer, co2 diffuses within ice cores.

  4. tchannon says:

    Gas is mobile with the atmosphere, can reflect quite differently from local conditions so there is no mystery provided in this case the ice stays ice.

    I’ve yet to see convincing evidence ice core soluble gas is a faithful record.

  5. oldbrew says:

    rishrac says: ‘Part of the rationale that the MWP never happened is that Antarctica didn’t melt and lead to much higher sea levels.’

    The MWP may have raised temps by a few degrees at most.

    Average Antarctic temp is minus several dozen. How much ‘extra’ melt would be likely?
    Remember sea ice melt has no effect on sea level.

  6. michael hart says:

    “justification to truncate at 7000 bp bad linearity”

    Sums up my suspicions about not just climate ice-core data, but climate tree-core data, and, everything else climate-core-proxy.

    Many (most?), but not all, phenomena in the universe will display an approximate first-order-kinetics exponential decay in their respective effects over time. Time and the universe does not generally afford the scientist a linear calibration curve. To apparently expect it, or assume it, is either naive, incompetent, or crooked.

  7. tchannon says:

    Completely agree Michael.

    Ice cores are particularly dubious, sucking on egos, the intense desire for deepest, oldest and so on. We’ll fix up time later.

    In this case, Taylor Dome, I think there is good evidence staring at us, core squeezes up then goes thump, solid at which point buckling, skewing, etc. does happen.

    But there is kidding on.

    According the these cores a solid is then compressed to much less than half it’s thickess, really, ever seen such a thing? Look at figure T12, what was it a bleed’in drought for tends of thousands of years? Of course not, the deeper cores are churned up junk.

    Here is probably the hallmark of a valid core, there is a definite rational timeline. Whether this reflects much on soluble gases, ah well.

  8. ren says:

    Area strongest radiation GCR shows the lowest pressure area over the Arctic Circle. It is obvious that over the Europe will be flow air from Greenland. In late November winter will attack in Europe.