Michael Asten: Paleo sensitivity study finds a third of IPCC central estimate

Posted: December 6, 2012 by Rog Tallbloke in Analysis, atmosphere, Carbon cycle, climate, Dataset, Ice ages

The new open peer review system on clim-past.net seems to be off to a good start with responses to this discussion paper by Michael Asten of Monash University:

Estimate of climate sensitivity from
carbonate microfossils dated near the
Eocene-Oligocene global cooling

M. W. Asten
School of Geosciences, Monash University, Melbourne, VIC 3800, Australia
Received: 9 September 2012 – Accepted: 17 September 2012 – Published: 5 October 2012
Correspondence to: M. W. Asten (michael.asten(at)monash.edu)
Published by Copernicus Publications on behalf of the European Geosciences Union.


Climate sensitivity is a crucial parameter in global temperature modelling. An estimate
is made at the time 33.4Ma using published high-resolution deep-sea temperature
proxy obtained from foraminiferal 18O records from DSDP site 744, combined with
published 5 data for atmospheric partial pressure of CO2 (pCO2) from carbonate microfossils,
where 11B provides a proxy for pCO2. The pCO2 data shows a pCO2 decrease
accompanying the major cooling event of about 4 C from greenhouse conditions to icecap
conditions following the Eocene-Oligocene boundary (33.7 My). During the cooling
pCO2 fell from 1150 to 770 ppmv. The cooling event was followed by a rapid and huge
10 increase in pCO2 back to 1130 ppmv in the space of 50 000 yr. The large pCO2 increase
was accompanied by a small deep-ocean temperature increase estimated as
0.59±0.063 C. Climate sensitivity estimated from the latter is 1.1±0.4 C (66% confidence)
compared with the IPCC central value of 3 C.

The post Eocene-Oligocene
transition (33.4 Ma) value of 1.1 C obtained here is lower than those published from
15 Holocene and Pleistocene glaciation-related temperature data (800 Kya to present) but
is of similar order to sensitivity estimates published from satellite observations of tropospheric
and sea-surface temperature variations. The value of 1.1 C is grossly different
from estimates up to 9 C published from paleo-temperature studies of Pliocene (3
to 4 Mya) age sediments. The range of apparent climate sensitivity values available
20 from paleo-temperature data suggests that either feedback mechanisms vary widely
for the different measurement conditions, or additional factors beyond currently used
feedbacks are affecting global temperature-CO2 relationships.


The paper has had responses, see the discussion here:


  1. Entropic man says:

    Note this short comment from one of the authors of a paper whose data was used by Asten.


    Asten would appear to be overreaching the limits of his data, giving a sensitivity of 1.1C+/-0.4C from data which could only support a range between 0.6C and 5.0C.

    Asten replies here: http://www.clim-past-discuss.net/8/C1910/2012/cpd-8-C1910-2012.pdf
    “On the problem of quantifying the change in pCO2, Pearson et al (2012) write “the
    difference in CO2 (delta-pCO2) between the rebound and the surrounding values (using
    the data tabulated in our Figure 1) is from as little as 77 to as much as 638 ppm.”
    Pearson et al make two quantitative mistakes in their comparison here while admitting
    that their calculation is just an exercise, not a serious estimate. Firstly they use 95%
    confidence limits, whereas my calculations use 66% confidence limits (as do the IPCC
    and the majority of referenced papers since 2007 when tabulating values for climate
    sensitivity). Secondly they compute baseline and higher values by addition and
    subtraction of the 2-sigma values for an individual point. Even though we have a limited
    data set it seems sensible to use statistical formulas for reduction of variance when
    averaging two numbers (whether for baseline or higher value) and addition of variance
    when differencing the higher value and baseline value.”

  2. oldbrew says:

    What about the time lag question?


    ‘The main evidence proving that CO2 does not control the climate, but at most can play a second fiddle by just amplifying the variations already present, is that of lags. In all cases where there is a good enough resolution, one finds that the CO2 lags behind the temperature by typically several hundred to a thousand years. Namely, the basic climate driver which controls the temperature cannot be that of CO2′

  3. michael hart says:

    Let’s not beat about the bush. By the definitions I have seen, “Climate-Sensitivity” means “Temperature-Sensitivity.” But Temperature is an intensive, not extensive, property. The term “global temperature” is a non-physical quantity, and so is a first-derivative thereof.

    Essex, McKitrick and Andresen authored on this matter. Non-paywalled versions are easy to find:

    I regard the matter largely as a “framing” debate, almost a side-issue. As the “temperature” fails to meet earlier projections/predictions, then proponents retrench to arguing about how fast it should be rising if only those pesky weather-fluctuations weren’t masking the expected catastrophe.

  4. Roger Andrews says:

    The graph below plots 80 estimates of climate sensitivity made since Arrhenius started the ball rolling in 1896. (Only mean or median estimates are plotted. The scatter would be a lot greater if the high and low estimates were included.)


    According to the graph we are no closer to obtaining a hard number for climate sensitivity now than we were a hundred years ago. The standard deviations show that climate sensitivity estimates have in fact been diverging, not converging, since the 1990s.

    I think the time has come to abandon efforts to quantify this unquantifiable – and arguably meaningless – metric and concentrate instead on figuring out how the earth’s climate works.

  5. michael hart says:

    “What about the time lag question?”

    Yes. Possibly the biggest problem of all for the “consensus” view. Papers like that of Shakun 2012 in Nature go into contortions attempting to wriggle round the issue and re-write the Paleo-record. Discussed in several places. E.g. http://wattsupwiththat.com/2012/04/08/did-shakun-et-al-really-prove-that-co2-precede-late-glacial-warming-part-1/

    I don’t buy it. Temperature leads CO2 in the modern (Mauna Loa) record too, and all that melting ice could dissolve a shed-load of atmospheric CO2 (solubility difference approximately 100-fold greater in water than ice). I sometimes wonder if IPCC proponents think that Chemistry doesn’t exist.

  6. Entropic man says:


    If 1 sigma or 66% confidence limits are becoming widely used, I consider either a mistake.I would prefer to see continued use of 95% confidence limits as standard.

    As for variance, I’ve been criticised by sceptics elsewhere for using variance adjusted HadCrut data, on the grounds that it has been devalued thereby. It’s a discussion I’ll leave to professionals.

    Regarding the overall issue of climate sensitivity, those competent to do the calculation, sceptic or acceptor, agree on a direct sensitivity around 1.1C per doubling. The sceptics cluster around 0.4C secondary forcing and the rest around 1.9C, but the uncertainties remain considerable. I notice the similarity between Aston’s 1.1+/-0.8C and Idso’s 1.1+0.4C. I tried a calclation once myself, based on warming rates and CO2 increase to date, and got 2.9C overall forcing.
    We can estimate and calculate until we’re blue in the face. Ultimately we’ll have to do the experiment; let the politicians run CO2 up to 560ppm, which looks likely after Rio and Doha, and see what happens.

    [Reply] Welcome to the uncertain world of climate science ‘confidence intervals’ :)

  7. Doug Proctor says:

    Sometimes the extended discussion makes me wonder about the initial position.

    As I understand the argument, GHGs work

    a) not by increasing the energy moving into a system, but

    2) by slowing down the process of release the energy, such that retention of thermal energy initially

    3) decreases overall radiative loss. The energy of the system then increases, and causes

    4) the temperature of the system to rise , until

    5) the increased temperature, through the 4th power of radiation release law, increases the rate of heat release of the subject material, unitl

    6) the rate of energy coming into the system is balanced by the rate of energy leaving the system, and the system is now stable, the only thing being different is that

    7) the temperature of the subject material is higher than previously.

    There must be some reason that bell jars filled with varying amounts of CO2 and water vapour with a constant, solar-equivalent energy source, in a non-variant outside-of-bottle atmospheric heat-dump environment can’t give us good radiative forcing numbers.


  8. Entropic man says:


    “[Reply] Welcome to the uncertain world of climate science ‘confidence intervals’”

    I was once trained in ecology. I’ve been used to marginal confidence levels in experimental data all my life. Even so, I still prefer 95% confidence limits.
    Dont you fall into the two value logic trap of assuming that because its measurement uncertainties cannot be precisely quantified, an effect does not exist.

    Doug Proctor

    “There must be some reason that bell jars filled with varying amounts of CO2 and water vapour with a constant, solar-equivalent energy source, in a non-variant outside-of-bottle atmospheric heat-dump environment can’t give us good radiative forcing numbers.”

    Visible light isnt absorbed by water vapour or CO2, so it goes straight through without warming the interior of your bell jar.
    The experiment works better if you have a solid rod such as a length of black-painted broomstick in the centre of the bell jar to absorb the visible light, act as a reservoir for the retained heat and emit IR. The CO2 and water vapour can then behave as GHGs do in an atmosphere.
    Set up like this, you find that jars with higher concentrations of GHGs heat up faster.

    [Reply] If the co2 concentration rose before the temperature rather than a couple of thousand years after, I’d take a bit more notice.

  9. Entropic man says:

    michael hart

    “”Climate-Sensitivity” means “Temperature-Sensitivity.””

    You are correct. Extra surface heating may come from downwelling IR from GHGs, increased solar insolation, urban heat islands, reduced cloud cover due to cosmic rays, ENSO, AMO or other sources.

    Whatever the primary source of the heating, secondary effects in terms of convection, evaporation, increased RH, cloud cover, increased decay of organic material, etc would ensue. These secondary forcings would probably be similar whateverer the cause of the initial heating.

    For example. the increased solar insolation observed by NASA from 1979 produced an increase of about 0.4W/M^2 at ground level, spread over 30 years. This would produce a direct temperature change of 0.11C.

    It would be interesting to know, at least approximately, what was the total effect of increased insolation on temperatures once secondary forcing was added in.

  10. Entropic man says:


    “[Reply] If the co2 concentration rose before the temperature rather than a couple of thousand years after, I’d take a bit more notice.”

    Try the bell jar experiment with my modification before you scoff. Even a simple two jar version with a bit of sodium bicarbonate and vinegar in the bottom of one, (and vinegar alone in the other) is enough. I did it in school with 2 litre plastic coke bottles, fluorescent lights and wooden dowels. Overnight they would generate a 2C temperature difference in favour of the bottle with extra CO2.

    [reply] Do you have photos of the experiments? Or can you do some diagrams? Do a write up for the blog.

  11. Coldish says:

    Thanks, Tallbloke, for showing us this example of the open review system in action. In view of some of the review comments, I wonder whether the author will make any modifications.
    If I wasn’t something of a climate sceptic already, a glance at this paper (and at some of the reviews) would have aroused my scepticism. The author uses a single deep sea site to estimate global paleotemperature (makes one-tree Briffa look almost rigorous by comparison) and relies on a proxy of uncertain reliability for CO2 from a different stratigraphic succession to come up with a figure for global response of temperature to atmospheric CO2 which he then expects the reader to take seriously.

  12. Entropic man says:

    I’m 12 years retired, but it should be easy enough to put an apparatus together. I’ll get back to you.

    [Reply] Great! We love real experiments here at the talkshop. If the bottles are sealed, don’t forget to put an accurate means of measuring pressure in them. That’s where similar previous experiments have gone wrong.