A NY Times science story (by Justin Gillis, March 1) illustrates some interesting points about science journalism–especially in the contentious and politically charged issue of climate change. A scientific journal, in this case the renowned Science magazine, wants to gain publicity for its journal among the non-scientific readers of the NY Times. So it sends out a press release about a scientific paper it plans to publish. Of course, it has been “peer-reviewed” to the satisfaction of the Science editor, but it has not yet been vetted by the scientific community.

Journalists who receive the press release think it will make an interesting story. It appears to involve some new features that contradict what has been discussed up till now. And Science has a generally good reputation. Specifically, the story claims that the deglaciation of 12,000 years ago, the major temperature increase marking the end of the most recent ice age and the beginning of the current warm Interglacial (termed The Holocene), involves also a striking increase in the greenhouse (GH) gas carbon dioxide. Investigators who wrote the paper seem to believe that the rise in carbon dioxide is coincident with the rise in temperature, as shown by their refined studies of ice cores from Antarctica.

This result has an important history. It starts with Al Gore announcing a correlation of several sudden temperature rises and carbon-dioxide increases during the recent ice age, as judged from analysis of Antarctic ice cores. He of course declares that this proves that CO2 has caused 20th-century warming. To his great embarrassment it was then later discovered that the increase in carbon dioxide actually follows the temperature increase by about 600-800 years. And even a non-scientist must realize that the cause must always precede the effect: so the temperature increase must be the cause of the carbon-dioxide increase—and not the other way round. The mechanism is really quite simple: When the ocean warms, it releases much of its dissolved carbon dioxide—similar to warming soda pop or champagne releasing CO2 bubbles. So suddenly, the “smoking gun” that Al Gore has relied on collapses in a heap.

This makes the new result interesting. The authors actually suggest that the carbon dioxide increase played a large role in the temperature increase terminating the last ice age. All of this sounds very good to the journalist who thinks he has a scoop—and to Science magazine that can get its name before the public.

Of course, the NYT editor also plays an important role. He realizes the political implications: If the scientific paper is correct, then perhaps Al Gore’s claim is vindicated and Congressman Joe Barton (R-TX), who had criticized Al Gore, is wrong. So the story is published before anyone has a chance to see the actual scientific paper and to know what the authors have discovered.

Please note that I’m not suggesting that any of the actual measurements are wrong. As reported, the work was done by a very competent group (in Grenoble, France), which has an excellent reputation on ice-core measurements. It is the interpretation of their findings that can be questioned. And beyond this, it illustrates how science can be used or misused in subtle ways to influence public opinion in directions governed by editorial policy.

Note also that the paper has certainly been “peer reviewed” by Science before being accepted for publication. But all this means is that the journal editor sent it to someone he chose and who will remain anonymous; and that this “someone” gave him a favorable report. If the editor had wanted to, he could have sent the paper to another reviewer, perhaps to someone who would have been critical about the paper’s conclusions. The public seems not to realize this discretionary power of a journal editor, and how little the label “peer review” really means.

It is customary in stories like this one for reporters to get the views of other experts, especially those who might disagree. There is no indication that this happened in this particular case. If Justin Gillis had phoned me for my reaction, I would have asked first of all to see the paper before giving an opinion. But even without seeing the paper, I would ask: Do the authors list the work of the Norwegian geologist Tom Segalstad, who described some problems in dating carbon dioxide found in ice cores. He and his co-author, the late Polish scientist Zbigniew Jaworowski, showed that the errors can be quite large. I would also have asked whether the authors list the relevant work of another Norwegian, Gunnar Myhre, who showed the “logarithmic” greenhouse effect of carbon dioxide: i.e., rather than proportional, the incremental effect diminishes as the amount of CO2 increases. This may mean that even if CO2 was responsible to some extent in aiding the ancient warming, the rapid deglaciation at the end of the last ice age, its effect may not be relevant to today’s situation, where the level of CO2 in the atmosphere is more than double the level during the last ice age.

Finally, I would ask the journalist whether he shouldn’t have mentioned the very pertinent fact that there hasn’t been any global warming for at least a decade, while carbon dioxide has been increasing rapidly—in large part because of the construction of so many coal-fired power plants in China.

What I’m suggesting therefore is that the story is unbalanced, one-sided, and should have been edited more severely. By the time the ice-core paper appears in Science magazine, all these concerns will have been forgotten—although as the scientific community studies the published work in detail, other problems may pop up.

Just to give one example: Could the Arctic have commenced the process of deglaciation a little ahead of the Antarctic? Might that account for a slightly earlier release of CO2 from a warming ocean? It’s a “chicken-and-egg” situation that needs to be looked at more carefully before jumping to facile conclusions. But for the NYT, politics trumps science.