New paper on climate sensitivity estimates 1.1 ± 0.4 °C for a doubling of CO2

This new paper (in review at the discussions section) at Climate of the Past has some interesting approaches  using

Oxygen 18 isotope records from benthic foraminiferas acquired in Deep Sea Drilling project (DSDP) on the Kerguelen Plateau off the coast of Antarctica and in the Cape Basin off the coast of Namibia. These drill holes provide

18O records with a resolution of order 10 000 yr across the Eocene-Oligocene boundary thus providing an excellent proxy for deep-ocean temperature.

Tbale 2 from the paper. Some estimates of climate sensitivity published 2004–2012. Values are in units

C, for a doubling of pCO2. Where published values are in units

C/(Wm−2), the published value is multiplied by 3.7 for the purpose of this comparison. Note that Asten’s median value of 1.1 agrees with Douglas and Christy.

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


Climate sensitivity is a crucial parameter in global temperature modelling. An estimate is made at the time 33.4 Ma using published high-resolution deep-sea temperature proxy obtained from foraminiferal δ18O records from DSDP site 744, combined with published data for atmospheric partial pressure of CO2 (pCO2) from carbonate microfossils, where δ11B 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 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 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 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.

Discussion Paper (PDF, 1101 KB)   Interactive Discussion (Open)

Readers that have access to Climate of the Past can leave a short comment until 30 Nov 2012. You can also watch the open review process as editors and reviewers leave comments. Constructive comments are welcome.


newest oldest most voted
Notify of

I may have read this abstract too quickly, but to me the term “climate sensitivity” assumes atmospheric CO2 drives global temperature, whereas the data I am familiar with clearly shows that CO2 lags temperature and is driven at least in part by temperature.


ooh, 1.1C (+/) will put quite a damper on Alarmist frenzies…..not much of a pretext for panic there…. they will need to find ways to discredit or ignore this paper!
p.s. typo alert for “Table 2” which reads “Tbale 2 from the paper”


When the climate cooled, more CO2 dissolved in the world’s oceans. When warmth returned so did the CO2.


Based on the well established metric that the small the number of authors, the more important the paper, this result is significant.

Steve from Rockwood

I have a problem with this “doubling” thing. CO2 levels haven’t doubled for tens (maybe hundreds) of thousands of years so why do we discuss the impossible?


I found this relevant to the post, so here ya go!
If you have the time, you will find it quite interesting, let alone the series…….

I think the earth will warm by a qua-jillion degrees Centigrade.

Dr Burns

Haven’t they heard that it has been shown that CO2 changes follow temperature over these periods ?

D Böehm

Dr Burns,
Correct. Which completely deconstructs the CO2=CAGW nonsense. ΔT causes ΔCO2, not vice versa — at least not that we can measure.


Dr Burns says:
October 5, 2012 at 8:20 pm
Haven’t they heard that it has been shown that CO2 changes follow temperature over these periods ?
Of course they have,but why let proven facts get in the way of a good crisis?

One thing is very interesting. This is very, very close to the best estimate of the date of inception of the Antarctic Circumpolar Current (opening/deepening of Drake Channel) at 33.6 My.
One could assume this produced a massive increase in Southern Ocean cyanobacterial productivity as, for the first time, it allowed for the mixing of volcanic dust (Southern Andes), glacial particulate material (Antarctica, Patagonia) and desert dust (Australia, Southern Africa) to blow and fallout onto and circulate through, the entire Southern Ocean.
This in turn would have pumped a whole lot more DMS (dimethyl sulfide) and other cloud nuclei into the air over the entire Southern Ocean thereby producing a marked dampening of global climate sensitivity (which persists to this day).
Personally, I’m getting real old just waiting for Barack Obama to put his jive mouth where our money is, bring one or two of those 100s of B-57 left over from the Cold War out of mothballs and top dress a big swathe of ocean with cheap (now the Chinese don’t want it) Aussie iron ore to show he really is the miracle man with the answer to Life, the Universe and damn well everything, (and the good ol’ US of A really does know how to lick that pesky CAGW).

Brian H

Since it’s both negative and imaginary, would that mean all CO2 sensitivity numbers should be multiplied by -1, or by its square root? Or both?


Why is there a semi-quaver preceding Oxygen 18?
[Reply: The HTML for Δ is “&+Delta+;”. But if you forget to capitalize Delta, you get: δ. — mod.]


Gee, that’s awfully close to the calculable and measurable Planck response of the climate to CO2.
1.23 degrees per doubling.

Sorry – typo in my post. I really meant B-52s not B-57. And the more I think about it , that would be a truly spectacular way for Barack Obama to demonstrate he isn’t just a One Trick Pony.

F. Ross

SamG says:
October 5, 2012 at 9:12 pm
Why is there a semi-quaver preceding Oxygen 18?

I believe this symbol is used to indicate a particular isotope of oxygen [or any other element]


The way the world’s economies are stagnating, we’ll be lucky to keep the plant-benefiting, biosphere-enriching increase in CO2 we’ve already got. Getting anywhere near a “doubling” will be next to impossible.

Jeff L

My recollection is that 1.1 ° C/ doubling sensitivity is actually close to John Christie’s estimate based on modern data & actually fits the current observed warming fairly well. Can anyone confirm that? Potentially another body blow to the scientific arguments of the alarmists, not that it will stop them from their alarming claims.

From the paper linked to above:

“…That separation is easier when a change in atmospheric CO2 concentration
can be shown to occur in a (geologically) short period of time since this gives opportunity
to observe the impulse response of temperature change associated with a rapid
atmospheric CO2 change. A coincidence of paleo-temperature and CO2 variations in
the geological record does not necessarily prove cause and effect, and unquestionably
10 the relationship will be multifactorial, however any observed relationship does provide
some bounds on possible models for the relationship

Some thoughts:
Better structured and organized than some of the other papers we’ve tried to look over.
Wording is still typical loose climatespeak with personal affections expressed that, in this case, muddy otherwise reasonably clear statements.
An apparent huge dependence on previous work casually aligning CO2 and Temperature lines and using their derived ratios to drive results in this paper. I didn’t see any attempts to verify pre-existing derived ratios prior to use.
As shown above, they admit coincidence. Then after admitting coincidence, they proceed to use the derived ratios to build their own graphs of coincidences.
Three separate data segments are defined on the charts (a,b, c). Each data segment appears to cover somewhat less than 90,000 years. Data segment ‘a’ provides; hole 744 – 12 points, hole 522 – Cibicidoides 4 points, Gyroidinoides 2 points. I assume this means 18 sections/slices of the core were tested. Furthermore, these specific slices were only tested for oxygen isotopes with only 18O isotope identified for us.
18 data points may make for a rough idea where the values are at points over 90,000 years. But I have difficulty accepting these data points in a discussion about temperature and CO2. It’s a shotgun pattern and they’re stating ‘error (1-sigma)’ results less than .023 average. ‘error (1-sigma)’ resullts for derived ‘change in temperature (deltaT)’ is .063. Which leaves me with the questions; OK, this is an accurate math result for change in temperature for a point in time, how is 90,000 years a point in time? And how does this make a reasonable coincidence with CO2?
CO2, according to the paper is an accepted pCO2 baseline from a Kilwa Formation coring… But I did not see any cross validation of holes 744, 522 to the Kilwa Formation core, especially for all points tested for this paper. How do I know they’re not comparing data points thousands of years apart?
While I think their conclusion sounds sensible; that does not make me feel comfortable that they’ve really done more than test two cores for oxygen isotopes. Calling it an “Estimate of climate sensitivity…” makes this paper another climate research coincidence, long on maybe, short on definitely.


CO2 following temperature is a positive feedback. Not a very strong one, but there, none the less. Because of it the temperature will be higher no matter the driving force that causes CO2 to increase. But what really matters is what happens next, and there is no evidence in history that what happens next is runaway heating or cooling. The temperature cycles up and down for varying periods but has never run off the dial at either end.
So what does happen next, or at least has always happened next so far? Temperature increases or decreases stop growing in the direction they were and begin to retreat. This has been going on for billions of years. Nobody can claim to know why.


Prof Murry Salby’s talk to the Sydney institute covers this topic very well.


Why is a seismologist writing a paper on CO2 and climate sensitivity?

David Ball

dp says:
October 5, 2012 at 9:57 pm
“CO2 following temperature is a positive feedback. Not a very strong one, but there, none the less. Because of it the temperature will be higher no matter the driving force that causes CO2 to increase.”
Perhaps I am unclear, but could you explain what these three sentences mean. Thank you in advance.

Garry says:
October 5, 2012 at 10:06 pm
Why is a seismologist writing a paper on CO2 and climate sensitivity?
It’s ground breaking research. get it?


What I find interesting is the “confidence” part in this one table. What the heck? That’s not science. That’s rolling dice. Seems like a way out for them. If it doesn’t happen they can always say “look, we weren’t 100% confident about it anyway.” More proof that science has been corrupted.

To determine the sensitivity they first have to know the forcing. The only forcing they seem to have data on is CO2. What was happening with the sun, with volcanics? CO2 is a lagging indicator of temperature. What was driving the temperature change? It seems to me that the abstract should be talking at least as much about what they can’t say about sensitivity as what they can say.


A climate sensitivity of ~1degC is about what Knut Angstrom predicted in 1900, isn’t it?


Would love it to be 1.1 and think it is, but the average from all the papers in that table is 2.675. Isnt that the best number from this article?

son of mulder

“Garry says:
October 5, 2012 at 10:06 pm
Why is a seismologist writing a paper on CO2 and climate sensitivity?”
Because not all scientists live in bunkers.


Asten claims that Antarctica and South America were separated substantially enough for a circumpolar current to have been established by the time of his study period. While he is not very specific,it seems implicit that this was a deep separation. From a leaf through the many papers on the timing and effect of the opening of the Drake Passage and a current favoring of a later date for full functional opening of the DP,I can’t see how Asten can be so definitive.

Using annual values, the increase in global temperaute from 1850 to 2011 was 0.783 °C (HadCRU3T) and the increase in CO2equivalent was from 289.0 to 464.1 ppm (GISS). If all that increase was caused by CO2e this implies a sensitivity of:
Sensitivity = 0.783 * log(2)/log(464.1/289.0 )= 1.145 °C per doubling of CO2e.
This is close the estimates of other climate realists.
For more details see:


How can CO2 make anything like this difference? There is not enough in the atmosphere to trap the radiation. If they can think that one CO2 molecule among 2500 odd others can reflect any significant radiation back to the surface, then they obviously have not done the mathematics. CO2 concentration in the atmosphere is not enough to affect the properties of all the others around it, it cannot do anything because there is just not enough of it.
This is as stupid as pretending that CO2 in seawater can make it more acidic, especially when it is only at concentrations of 109 ppm. This is only one molecule per 9174, and that cannot exert any sort of pH effect on its own, and even if you doubled it, tripled it, or multiplied it by 10, it would still not be enough to exert any effect..
The theory that CO2 can trap radiation, or acidify sea water in the concentrations it is present in our atmosphere and oceans is physically impossible, and no amount of Mickey Mouse conclusions will change that


@Allan McRae
Climate sensitivity to CO2 is independent of how the CO2 got there. It’s based on radiative transfer properties and feedbacks. The reason for the great range of estimates of sensitivity is because the feedbacks are so uncertain. Doubling CO2 alone without accounting for feedbacks gives a warming effect of 1.1 C. So estimates close to that number assume feedbacks cancel out. The bigger issue is whether to regard CO2 as a forcing or a feedback.
If warming from unknown forcings cause more CO2 to dissolve out of the oceans, for example, then CO2 itself would be a feedback to that warming event, causing even more warming. But because CO2s warming effect is logarithmic, whilst warming events tend to be quite linear, it’s effect as a feedback is not linear. You need more and more CO2 to get the same warming effect.

SamG says:

Why is there a semi-quaver preceding Oxygen 18?

You might have the wrong character encoding. With the correct encoding (UTF-8) I see a delta sign, which you probably know means “change in ” 18O.

I can claim to know why! ( I can’t claim to know that I’m certainly correct, though 😉
At the hot end of things, evaporation and convection pick up faster and faster, winning the heating race and putting a lid on warming at about 100 F / 40 C (i.e. in the tropics you get lots of thunderstorms in the afternoon when it gets that hot).
At the cold end, ice covers the water and cold suppresses the heat transport. The convection evaporation cycle shuts down.
So it stops at one end, and “wins the race” at the other, as the evaporation / convection cycle varies faster with temperature than the actual heat change.
That’s how heat pipes work, and why they have prescribed ranges of operation.
On the Alaska Pipeline, there are heat pipes that cool the ground in winter, but do not warm it in summer. One way heat transfer out of the ground to assure it stays frozen. When the air end is hotter, it shuts down, when the air end is colder, it runs… until the temperature drops below the coolant evaporation point, then it stops again…

Garry says:

Why is a seismologist writing a paper on CO2 and climate sensitivity?

Prof Asten is a geophysicist, not a seismologist. See
I met him in the early 1990s doing mining exploration, and I remember him as a very capable individual. He has had a bit of flack from the Aus media as an “evil denier”, but I believe that reflects on the media, more than on Prof Asten.

cui bono

SamG says: Why is there a semi-quaver preceding Oxygen 18?
’tis the Music of the Spheres….

It is not temperature that is important but heat content which varies at a given temperature with water content. It is heat that drives weather/climate. Whilst 18O proxy data is good for temperature it does not relate to heat content.
Go back and recalculate.


(sort of off-topic)
Anthony~ While the bulk of your article is pretty much taken straight from the paper mentioned, this is still Your article. I’d like to repost it to the site I lay out my skeptic arguments at, if I may. (with a link back here, of course)
May as well say it here, since many of the people I would like to ask, come by here:
I am a Skeptic in a sea of True Believers, at an art site known as Deviantart (.com). I’ve written a number of articles on climate change there, and have hopes to do a great many more.
It goes Much faster, if I can repost articles written by others, along with my own commentary / thoughts on the matter.
Don’t know that I will get many responses here (or if this comment will even be published), but thought I would give it a try.

This article is not a paper in Climate of the Past. It is in Climate of the Past Discussion. It is not peer reviewed. The EGU journals have a unique open peer review process, where submitted manuscripts that pass a quality control are placed immediately online for the community to read and comment on. Articles that pass peer review are then published in Climate of the Past. CoPD is more akin to a preprint server like than a standard journal.
A huge advantage of submitting to CoPD is that the community can read your ideas almost immediately. The great disadvantage is that any embarrassing mistakes you make in your submission are preserved for the public in perpetuity rather than remaining a secret of the editor and reviewers.
Asten’s article makes some heroic assumptions to calculate this low value of climate sensitivity. Lets just look at one. Asten (2012) scales the isotope-inferred temperature change in the deep ocean by 1.5 to get global mean temperature change. This scale factor comes from Hansen and Sato (2012) who calculated it by comparing a global compilation of deep ocean oxygen isotope records with estimated global temperature over the Pleistocene. There are three major problems here. First, Asten (2012) assumes that this scale factor is without any uncertainty. Allowing for the inherent uncertainty in the estimate from Hansen and Sato (2012) will increase the uncertainty in Asten’s (2012) result. Second, Asten (2012) assumes that a global average relationship between deep ocean temperature and surface temperatures is valid for any single location. This is extremely doubtful. Thirdly Asten (2012) assumes that a scaling factor that is valid for the Pleistocene is also valid for the Oligocene, with different ocean circulation.
REPLY: Ah old ‘Buzzkill Telford’, who never has anything positive to say here.
“It is not peer reviewed.” Perhaps you forgot to look at the end links and the note I left about leaving comments. But for the dour like yourself, I’ll make it clearer – Anthony


SamG says:
October 5, 2012 at 9:12 pm
Why is there a semi-quaver preceding Oxygen 18?

It’s supposed to be a lower case delta. The term is used to designate the 18O:16O ratio in a sample against the ratio in a standard.


Steve from Rockwood says:
October 5, 2012 at 8:09 pm
I have a problem with this “doubling” thing. CO2 levels haven’t doubled for tens (maybe hundreds) of thousands of years so why do we discuss the impossible?

Try thinking about it this way: If you want to increase the temperature 1.1 degrees, you have to double the CO2 in the atmosphere. If you want to increase the temperature 2.2 degrees, you have to quadruple the CO2 in the atmosphere. 3.3 degrees requires eight times the CO2 and so forth.


Just to clarify, this is a ‘paper under review’, right? It hasn’t been accepted for publication yet? If so, then obviously its conclusions need to be treated with a degree of caution.
Re post 1, Allan MacRae: can you confirm that you still believe a sixth order is “the best fit polynomial” trend line to use for the UAH data, as you stated in this Sept. 2008 article at ‘ICECAP’:

Do they or somebody also have a figure for how much cooling is caused (co2 has absorption in the uv and at 2 and 5 um.

Bill Illis

Technically, the data at this time period says the CO2 sensitivity is Null.
Global temps fall 2.0C starting 33.6 million years over a few hunded thousand years to a level close to today. Antarctica glaciates over in under 100,000 years. CO2 stays high at 1200 ppm over the period.
Temps stay more-or-less at this level for the next 7 million years. CO2 falls temporarily, 2 million years later, but then it goes back up again.
The Antarctic Circumpolar Current is constricted again at 27 million years ago and the glaciers melt back and global temperatures rise 2.0C again. CO2 then goes down, just as the temperatures are going back up.
Temperatures are still warm at 24 million years ago, the Antarctic glaciers are only at about half of today’s level. CO2 promptly falls below 280 ppm for the perhaps the very first time in history. Temperatures stay high and even go higher over the next 8 million years while CO2 stays in a range of 240 ppm to 400 ppm.
There is no correlation. .
All of the CO2 sensitivity studies in the table above did not use the actual temperature and CO2 estimates. They more-or-less just made-up the values based on some climate model simulation or some backwards math technique. You cannot get a paper published that says there is no correlation between CO2 and temperature in X period (even though the data shows this).


‘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’
So he’s trying to redefine climate sensitivity? The deep-ocean temperature response to forcing changes?
I don’t think you’re allowed to do that…

Louis Hooffstetter

David Ball:
“CO2 following temperature is a positive feedback. Not a very strong one, but there, none the less. Because of it the temperature will be higher no matter the driving force that causes CO2 to increase.”
I believe DP means that as the oceans warm (regardless of why) they release CO2. This additional CO2 will have a slight positive feedback which will cause atmospheric temperatures to rise even more.

Doug Huffman

Heat content varies with the specific heat integrated across the stuff.
Like the move to simplified language standard English in the law, this abstract, particularly, is so full of jargon and acronym that it is nearly unreadable nonsense.

I see no one read the paper. See the scaling factors in eq 4.
REPLY: I see Mosh still can’t make complete comments. Steve when will you learn to be more effective by typing more than few crypto bits? It’s a chronic problem with you. I think of you like a brother in arms, but please if you want to make a point, make it with substance. Thanks – Anthony

“additional factors beyond currently used feedbacks are affecting global temperature-CO2 relationships.”
Let’s get it out.
H2O vapor/clouds makes it cooler when other factors cause warmth. H2Ovapor/clouds makes it warmer when other factors cause cooling.
This is my story and I’m sticking to it.