Pearson, P.N. and Palmer, M.R. 2000. Atmospheric carbon dioxide concentrations over the past 60 million years. Nature 406: 695-699.
What was done
The authors used boron-isotope ratios of ancient planktonic foraminifer shells to estimate the pH of surface-layer sea water throughout the past sixty million years, which they then used to reconstruct a history of atmospheric CO2 concentration over this period, which they finally compared with oxygen isotope ratios of deep sea benthic foraminifera that serve as proxies for temperature.
What was learned
Supposedly, good records of both atmospheric CO2 concentration and oxygen isotope values were obtained for the past 24 million years and for the period from 40 to 60 million years ago.
What it means
The authors state that "change in the carbon dioxide concentration of the atmosphere is commonly regarded as a likely forcing mechanism on global climate over geological time because of its large and predictable effect on temperature," which "predictable effect" is that increases in atmospheric CO2 concentration cause higher temperatures to occur and that decreases in atmospheric CO2 concentration cause lower temperatures to occur. Their data, however, clearly demonstrate that this incredibly common assumption is just plain false.
Starting 60 million years before present (BP), the authors have the atmosphere's CO2 concentration at approximately 3600 ppm and the oxygen isotope ratio at about 0.3 per mil. Thirteen million years later, however, the air's CO2 concentration has dropped all the way down to 500 ppm; but the oxygen isotope ratio has dropped (implying a rise in temperature) to zero, which is, of course, just the opposite of what one would expect from the "large and predictable effect" of CO2 on temperature that is commonly assumed.
Next comes a large spike in the air's CO2 content, all the way up to a value of 2400 ppm. And what does the oxygen isotope ratio do? It rises slightly (implying temperature falls slightly) to about 0.4 per mil, which is again just the opposite of what one would expect from the "large and predictable effect" of CO2 on temperature that is commonly assumed.
After the spike in CO2, of course, the air's CO2 concentration drops dramatically, declining to a minimum value of close to what it is today. And the oxygen isotope ratio? It barely changes at all, defying once again the common assumption of the "large and predictable effect" of CO2 on temperature.
Between this point and the break in the record at 40 million years BP, the air's CO2 concentration rises again to approximately 1000 ppm; and - need we say? - the oxygen isotope ratio rises slightly (implying a slight cooling) to 0.6 per mil. And once again, well, you get the picture: the common assumption fails miserably.
Picking up the record at 24 million years BP, there are but relatively tiny variations in atmospheric CO2 concentration up to the present; but, of course, there are large variations in oxygen isotope values, both up and down, again in clear contradiction of the "common assumption."
The most interesting of these last oxygen isotope changes is the dramatic increase (implying a dramatic cooling) over the most recent two million years, when, of course, the air's CO2 concentration has actually risen slightly.
Now you tell us that you still believe in CO2-induced global warming. And if you do, we'd like to talk to you about a bridge we have for sale.
Reviewed 20 September 2000